|Publication number||US7822197 B2|
|Application number||US 11/899,662|
|Publication date||Oct 26, 2010|
|Filing date||Sep 7, 2007|
|Priority date||Jul 2, 1996|
|Also published as||US7457962, US7647502, US7647503, US7664958, US7877609, US7930545, US7953981, US7991188, US8281140, US20070011458, US20070226506, US20070300072, US20070300073, US20080022113, US20080022114, US20080046742, US20100005308, US20100064140, US20100077220|
|Publication number||11899662, 899662, US 7822197 B2, US 7822197B2, US-B2-7822197, US7822197 B2, US7822197B2|
|Inventors||Scott A. Moskowitz|
|Original Assignee||Wistaria Trading, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (270), Non-Patent Citations (96), Referenced by (3), Classifications (8), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application is a divisional of U.S. patent application Ser. No. 09/789,711, filed Feb. 22, 2001, now U.S. Pat. No. 7,107,451, which is a continuation-in-part of U.S. patent application Ser. No. 09/281,279, filed Mar. 30, 1999, now U.S. Pat. No. 6,522,767, which is a continuation of U.S. patent application Ser. No. 08/677,435, filed Jul. 2, 1996, now U.S. Pat. No. 5,889,868. The previously identified patents and/or patent applications are hereby incorporated by reference, in their entireties.
This application claims the benefit of U.S. patent application Ser. No. 08/674,726, filed Jul. 2, 1996, entitled “Exchange Mechanisms for Digital Information Packages with Bandwidth Securitization, Multichannel Digital Watermarks, and Key Management” (issued as U.S. Pat. No. 7,362,775 on Apr. 22, 2008); pending U.S. patent application Ser. No. 08/999,766, filed Jul. 23, 1997, entitled “Steganographic Method and Device”; U.S. patent application Ser. No. 09/046,627, filed Mar. 24, 1998, entitled “Method for Combining Transfer Function with Predetermined Key Creation” (issued as U.S. Pat. No. 6,598,162); U.S. patent application Ser. No. 09/053,628, filed Apr. 2, 1998, entitled “Multiple Transform Utilization and Application for Secure Digital Watermarking” (issued as U.S. Pat. No. 6,205,249); U.S. patent application Ser. No. 09/281,279, filed Mar. 30, 1999, entitled “Optimization Methods for the Insertion, Protection, and Detection of Digital Watermarks in Digital Data” (issued as U.S. Pat. No. 6,522,767); U.S. Provisional Application No. 60/169,274, filed Dec. 7, 1999, entitled “Systems, Methods And Devices For Trusted Transactions” (U.S. patent application Ser. No. 09/731,040, which issued as U.S. Pat. No. 7,159,116 on Jan. 2, 2007); U.S. patent application Ser. No. 09/456,319, filed Dec. 8, 1999, entitled “Z-Transform Implementation of Digital Watermarks” (issued as U.S. Pat. No. 6,853,726); U.S. patent application Ser. No. 09/545,589, filed Apr. 7, 2000, entitled “Method and System for Digital Watermarking” (issued as U.S. Pat. No. 7,007,166); U.S. patent application Ser. No. 09/594,719, filed Jun. 16, 2000, entitled “Utilizing Data Reduction in Steganographic and Cryptographic Systems” (which is a continuation-in-part of International Application No. PCT/US00/06522, filed Mar. 14, 2000, which PCT application claimed priority to U.S. Provisional Application No. 60/125,990, filed Mar. 24, 1999) (issued as U.S. Pat. No. 7,123,718); International Application No. PCT/US00/21189, filed Aug. 4, 2000 (which claims priority to U.S. Patent Application No. 60/147,134, filed Aug. 4, 1999, and to U.S. Patent Application No. 60/213,489, filed Jun. 23, 2000, both of which are entitled, “A Secure Personal Content Server”) (issued as U.S. Pat. No. 7,475,246 on Jan. 6, 2009), U.S. patent application Ser. No. 09/657,181, filed Sep. 7, 2000, entitled “Method And Device For Monitoring And Analyzing Signals” (issued as U.S. Pat. No. 7,346,472 on Mar. 18, 2008); U.S. Provisional Patent Application No. 60/234,199, filed Sep. 20, 2000, entitled “Improved Security Based on Subliminal and Supraliminal Channels For Data Objects” (issued as U.S. Pat. No. 7,127,615); and U.S. patent application Ser. No. 09/671,739, filed Sep. 29, 2000, entitled “Method And Device For Monitoring And Analyzing Signals,” (abandoned) U.S. patent application Ser. No. 09/731,039 entitled “System and Method for Permitting Open Access to Data Objects and For Securing Data Within the Data Objects,” filed Dec. 7, 2000 (issued as U.S. Pat. No. 7,177,429); and U.S. patent application Ser. No. 09/731,040, entitled “Systems, Methods and Devices for Trusted Transactions,” filed Dec. 7, 2000 (issued as U.S. Pat. No. 7,159,116). The previously identified patents and/or patent applications are hereby incorporated by reference, in their entireties.
In addition, this application hereby incorporates by reference, as if fully stated herein, the disclosures of U.S. Pat. No. 5,613,004 “Steganographic Method and Device”; U.S. Pat. No. 5,745,569 “Method for Stega-Cipher Protection of Computer Code”; U.S. Pat. No. 5,889,868 “Optimization Methods for the Insertion, Protection, and Detection of Digital Watermarks in Digitized Data”; and U.S. Pat. No. 6,078,664, entitled “Z-Transform Implementation of Digital Watermarks.”
The present invention relates to digital watermarks.
Digital watermarks exist at a convergence point where creators and publishers of digitized multimedia content demand localized, secured identification and authentication of that content. Because existence of piracy is clearly a disincentive to the digital distribution of copyrighted works, establishment of responsibility for copies and derivative copies of such works is invaluable. In considering the various forms of multimedia content, whether “master,” stereo, NTSC video, audio tape or compact disc, tolerance of quality degradation will vary with individuals and affect the underlying commercial and aesthetic value of the content. It is desirable to tie copyrights, ownership rights, purchaser information or some combination of these and related data to the content in such a manner that the content must undergo damage, and therefore a reduction in value, with subsequent, unauthorized distribution of the content, whether it be commercial or otherwise.
Legal recognition and attitude shifts, which recognize the importance of digital watermarks as a necessary component of commercially distributed content (audio, video, game, etc.), will further the development of acceptable parameters for the exchange of such content by the various parties engaged in the commercial distribution of digital content. These parties may include artists, engineers, studios, Internet access providers, publishers, agents, on-line service providers, aggregators of content for various forms of delivery, on-line retailers, individuals and parties that participate in the transfer of funds to arbitrate the actual delivery of content to intended parties.
Since the characteristics of digital recordings vary widely, it is a worthwhile goal to provide tools to describe an optimized envelope of parameters for inserting, protecting and detecting digital watermarks in a given digitized sample (audio, video, virtual reality, etc.) stream. The optimization techniques described hereinafter make unauthorized removal of digital watermarks containing these parameters a significantly costly operation in terms of the absolute given projected economic gain from undetected commercial distribution. The optimization techniques, at the least, require significant damage to the content signal, as to make the unauthorized copy commercially worthless, if the digital watermark is removed, absent the use of extremely expensive tools.
Presumably, the commercial value of some works will dictate some level of piracy not detectable in practice and deemed “reasonable” by rights holders given the overall economic return. For example, there will always be fake $100 bills, LEVI'S® jeans, and Gucci® bags, given the sizes of the overall markets and potential economic returns for pirates in these markets—as there also will be unauthorized copies of works of music, operating systems (Windows® 95, etc.), video and future multimedia goods.
However, what differentiates the “digital marketplace” from the physical marketplace is the absence of any scheme that establishes responsibility and trust in the authenticity of goods. For physical products, corporations and governments mark the goods and monitor manufacturing capacity and sales to estimate loss from piracy. There also exist reinforcing mechanisms, including legal, electronic, and informational campaigns to better educate consumers.
According to one embodiment of the present invention, a system for encoding of digital watermark information in a signal is disclosed. The system includes a window identifier for identifying a sample window in the signal; an interval calculator for determining a quantization interval of the sample window; and a sampler for normalizing the sample window to provide normalized samples. In general, the sample window that is identified will have a maximum and a minimum. The quantization interval is used to quantize normalized window samples. The normalized samples conform to a limited range of values that are proportional to real sample values and comprise a representation of the real sample values with a resolution higher than the real range of values. The normalized values may also be divided by the quantization interval into distinct quantization levels.
According to another embodiment of the present invention, a system for pre-analyzing a digital signal for encoding at least one digital watermark using a digital filter is disclosed. The system includes a processor for identifying an area of the digital signal that will be affected by the digital filter and an encoder for encoding the at least one digital watermark in the digital signal. The encoder encodes the at least one digital watermark so as to avoid the at least one area of the digital signal that will be affected by the digital filter.
According to another embodiment of the present invention, a system for preprocessing a watermark message is disclosed. The system includes a pre-processor for determining an exact length of a watermark message as it will be encoded, and a key generator for generating a watermark key that provides at least one unique bit for each bit comprising the watermark message.
According to still another embodiment of the present invention, a system for encoding a watermark in a digital signal is disclosed. The system includes a generator for generating a plurality of watermark pseudo-random key bits, and an encoder for encoding the watermark in the digital signal using the watermark pseudo-random key bits and characteristics of the digital signal. The generator may be a non-linear generator, a chaotic generator, etc.
According to another embodiment of the present invention, a system for encoding a watermark in a digital signal is disclosed. The system includes a mapper for mapping pseudo-random key and processing state information to effect an encode/decode map using a generator, and an encoder for encoding the watermark in the digital signal using the encode/decode map and characteristics of the digital signal. The generator may be a non-linear generator, a chaotic generator, etc.
According to another embodiment of the present invention, a system for encoding watermarks is disclosed. The system includes an inverter for inverting at least one instance of the watermark bit stream, and an encoder for encoding at least one instance of the watermark using the inverted instance of the watermark bit stream.
According to another embodiment of the present invention, a system for analyzing composite digitized signals for watermarks is disclosed. The system includes a first receiver for receiving a composite signal; a second receiver for receiving an unwatermarked sample signal; an aligner for time aligning the unwatermarked sample signal with the composite signal; an adjuster for gain adjusting the time aligned unwatermarked sample signal to a corresponding segment of the composite signal, determined when the signals are time aligned; an estimator for estimating a pre-composite signal using the composite signal and the gain adjusted unwatermarked sample signal; an estimator for estimating a watermarked sample signal by subtracting the estimated pre-composite signal from the composite signal; and a scanner for scanning the estimated watermarked sample signal for watermarks.
According to another embodiment of the present invention, a method for pre-analyzing a digital signal for encoding a plurality of digital watermarks using a digital filter is disclosed. The method includes the steps of (1) providing a plurality of digital watermarks; (2) determining an encoding level; and encoding each of the plurality of digital watermarks at substantially the same encoding level.
According to another embodiment of the present invention, a method for pre-analyzing a digital signal for encoding digital watermarks using a digital filter is disclosed. The method includes the steps of (1) providing a digital signal; (2) providing a digital filter to be applied to the digital signal; and (3) identifying an area of the digital signal that will be affected by the digital filter based on at least one measurable difference between the digital signal and a counterpart of the digital signal selected from the group consisting of the digital signal as transmitted, the digital signal as stored in a medium, and the digital signal as played backed.
According to another embodiment of the present invention, a method for encoding a watermark in a content signal is disclosed. The method includes the steps of (1) splitting a watermark bit stream; and (2) encoding at least half of the watermark bit stream in the content signal using inverted instances of the watermark bit stream.
According to another embodiment of the present invention, a method for encoding at least one watermark in a content signal is disclosed. The method includes the steps of (1) predetermining a number of bits in the content signal to be encoded, based on at least one of a fixed length key and signal characteristics of the content signal; and (2) encoding the watermark in the predetermined bits.
According to another embodiment of the present invention, a method for encoding at least one watermark in a content signal is disclosed. The method includes the steps of (1) locating at least one noise-like signal feature in the content signal; and (2) encoding the at least one watermark in substantially the same location as the at least one noise-like signal feature.
According to another embodiment of the present invention, a method for encoding at least one digital watermark in a content signal is disclosed. The method includes the steps of (1) measuring a perceived signal-to-error ratio; and (2) encoding the at least one watermark in a channel bound by a minimum and maximum signal-to-error level for the content signal.
According to another embodiment of the present invention, a method for digital watermark encode/decode comprises the steps of: (1) measuring a perceived signal-to-error ratio; and (2) encoding at least one watermark in a signal feature that is bound by a minimum and maximum signal-to-error level for the digital signal.
According to another embodiment of the present invention, a method for decoding a digital watermark from a content signal is disclosed. The method includes the steps of (1) receiving a suspect digital signal to be analyzed; (2) subjecting the digital signal to a time-based alignment; (3) using the time-based alignment to align amplitude values in the suspect digital signal; and (4) decoding a digital watermark.
According to another embodiment of the present invention, a method for encoding watermarks in a content signal is disclosed. The method includes the steps of (1) identifying a plurality of signal features in the content signal; and (2) inserting watermark data in the signal features. The signal features may be identified from relationships between multiple sample windows in the content signal.
According to another embodiment of the present invention, a method for decoding watermarks in a content signal is disclosed. The method includes the steps of (1) identifying a plurality of signal features in the content signal; and (2) decoding watermark data from the signal features. The signal features may be identified from relationships between multiple sample windows in the content signal.
According to another embodiment of the present invention, a method for pre-analyzing a digital signal for encoding digital watermarks using a digital filter is disclosed. The method includes the steps of (1) identifying at least one of a frequency and a time delimited area of the signal that will be affected by the digital filter; and (2) encoding at least one digital watermark so as to avoid the identified area.
According to another embodiment of the present invention, a method for pre-analyzing a digital signal for encoding digital watermarks using a digital filter is disclosed. The method includes the steps of (1) identifying at least one change to the digital signal that will be affected by the digital filter; and (2) encoding at least one digital watermark so the watermark survives the changes introduced by the digital filter.
According to another embodiment of the present invention, a method for guaranteeing watermark uniqueness is disclosed. The method includes the steps of (1) providing a watermark; and (2) attaching a timestamp to the watermark.
According to another embodiment of the present invention, a method for guaranteeing watermark uniqueness is disclosed. The method includes the steps of (1) providing a watermark; and (2) attaching a user identification dependent hash to the watermark.
According to another embodiment of the present invention, a method for guaranteeing watermark uniqueness is disclosed. The method includes the steps of (1) providing a watermark; and attaching a message digest of watermark data to the watermark.
According to another embodiment of the present invention, a system for digital watermark encode/decode operations comprises: (1) a CODEC database comprising a plurality of CODECs; and (2) a processor which encodes at least one watermark using at least one CODECs from the CODEC database.
According to another embodiment of the present invention, a method for digital watermark encode/decode is disclosed. The method includes the steps of (1) receiving a digital signal stream; (2) using one or more of a plurality of watermarking CODECs; and (3) encoding/decoding at least one of a digital watermark and associating one or more of a plurality of watermarking CODECs with a predetermined key.
According to another embodiment of the present invention, an article of manufacture comprises: a receiver to receive a digital signal; a detector to detect at least two of a plurality of digital watermarks located within the digital signal; and a processor that enables content signal manipulation of the digital signal based on successful detection of at least two of the plurality of digital watermarks. The article may also include a verification module which verifies at least one detected watermark.
According to another embodiment of the present invention, a method for preprocessing a digital data signal to authorize a plurality of unique descendant copies of the digital data signal is disclosed. The method includes the steps of (1) providing a digital data signal; (2) identifying candidate bits of the digital data signal that will be manipulated during embedding; (3) generating a digital watermark message to be embedded based on at least one predetermined criterion; and (4) embedding the digital watermark message in the digital data signal. The candidate bits may be identified through a psychoacoustic or psychovisual model.
According to another embodiment of the present invention, a method for preprocessing a digital data signal to authorize a plurality of unique descendant copies of the digital data signal is disclosed. The method includes the steps of (1) providing a digital data signal; (2) identifying candidate bits of the digital data signal that will be manipulated during scrambling; (3) generating a scrambling key on at least one predetermined criterion; and (4) scrambling the digital data signal with the scrambling key. The candidate bits may be identified through a psychoacoustic or psychovisual model.
According to another embodiment of the present invention, a method for creating a descendant copy of a digital data signal is disclosed. The method includes the steps of (1) obtaining a model for the digital data signal; and (2) generating a watermark for the descendant copy of the digital data signal based on at least one criterion. The criterion may be, for example, a geographical territory, a transaction identification, an individual identification, a use limitation, a signal domain, etc.
The present invention relates to implementations of digital watermarks that are optimally suited to particular transmission, distribution and storage mediums given the nature of digitally sampled audio, video, and other multimedia works.
The present invention also relates to adapting watermark application parameters to the individual characteristics of a given digital sample stream.
The present invention additionally relates to the implementation of digital watermarks that are feature-based. That is, a system where watermark information is not carried in individual samples, but is carried in the relationships between multiple samples, such as in a waveform shape. For example, in a manner similar to the way a US $100 bill has copy protection features including ink type, paper stock, fiber, angles of artwork that distort in photocopier machines, inserted magnetic strips, and composite art, the present invention envisions natural extensions for digital watermarks that may also separate frequencies (color or audio), channels in 3D while utilizing discreteness in feature-based encoding only known to those with pseudo-random keys (i.e., cryptographic keys) or possibly tools to access such information, which may one day exist on a quantum level.
There are a number of hardware and software approaches in the prior art that attempt to provide protection of multimedia content, including encryption, cryptographic containers, cryptographic envelopes or “cryptolopes,” and trusted systems in general. None of these systems places control of copy protection in the hands of the content creator as the content is created, nor provides an economically feasible model for exchanging the content to be exchanged with identification data embedded within the content.
Yet, given the existence of over 100 million personal computers and many more non-copy-protected consumer electronic goods, copy protection seems to belong within the signals. After all, the playing (i.e., using) of the content establishes its commercial value.
Generally, encryption and cryptographic containers serve copyright holders as a means to protect data in transit between a publisher or distributor and the purchaser of the data (i.e., a means of securing the delivery of copyrighted material from one location to another by using variations of public key cryptography or other more centralized cryptosystems).
Cryptolopes are suited specifically for copyrighted text that is time-sensitive, such as newspapers, where intellectual property rights and origin data are made a permanent part of the file. For information on public-key cryptosystems see U.S. Pat. No. 4,200,770 to Hellman et al., U.S. Pat. No. 4,218,582 to Hellman et al., U.S. Pat. No. 4,405,829 to Rivest et al., and U.S. Pat. No. 4,424,414 to Hellman et al., which patents are incorporated herein by reference. Systems are proposed by IBM and Electronic Publishing Resources to accomplish cryptographic container security.
Digitally-sampled copyrighted material, that is binary data on a fundamental level, is a special case because of its long term value coupled with the ease and perfection of copying and transmission by general purpose computing and telecommunications devices. In particular, in digitally-sampled material, there is no loss of quality in copies and no identifiable differences between one copy and any other subsequent copy. For creators of content, distribution costs may be minimized with electronic transmission of copyrighted works. Unfortunately, seeking some form of informational or commercial return via electronic exchange is ill-advised absent the use of digital watermarks to establish responsibility for specific copies and unauthorized copying. Absent digital watermarks, the unlikely instance of a market of trusted parties who report any distribution or exchange of unauthorized copies of the protected work must be relied upon for enforcement. Simply, content creators still cannot independently verify watermarks should they choose to do so.
For a discussion of systems that are oriented around content-based addresses and directories, see U.S. Pat. No. 5,428,606 to Moskowitz, which patent is incorporated herein by reference.
In combining steganographic methods for insertion of information identifying the title, copyright holder, pricing, distribution path, licensed owner of a particular copy, or a myriad of other related information, with pseudo-random keys (which map insertion location of the information) similar to those used in cryptographic applications, randomly placed signals (digital watermarks) can be encoded as random noise in a content signal. Optimal planning of digital watermark insertion can be based on the inversion of optimal digital filters to establish or map areas comprising a given content signal insertion envelope. Taken further, planning operations will vary for different digitized content: audio, video, multimedia, virtual reality, etc. Optimization techniques for processes are described in related patents, including U.S. Pat. No. 5,613,004, entitled “Steganographic Method and Device” and U.S. Pat. No. 5,822,432, entitled “Method for Human Assisted Random Key Generation and Application for Digital Watermark System,” which patents are incorporated herein by reference.
Optimization processes must take into consideration the general art of digitization systems where sampling and quantizing are fundamental physical parameters. For instance, discrete time sampling has a natural limit if packets of time are used, estimated at 1.times.10.sup.−42 second. This provides a natural limit to the sampling operation. Also, since noise is preferable to distortion, quantizing will vary given different storage mediums (magnetic, optical, etc.) or transmission mediums (copper, fiber optic, satellite, etc.) for given digitized samples (audio, video, etc.). Reducing random bit error, quantization error, burst error, and the like is done for the singular goal of preserving quality in a given digitized sample. Theoretical perfect error correction is not efficient, given the requirement of a huge allocation of redundant data to detect and correct errors. In the absence of such overhead, all error correction is still based on data redundancy and requires the following operations: error detection to check data validity, error correction to replace erroneous data, and error concealment to hide large errors or substitute data for insufficient data correction. Even with perfect error correction, the goal of a workable digital watermark system for the protection of copyrights would be to distribute copies that are less than perfect but not perceivably different from the original. Ironically, in the present distribution of multimedia, this is the approach taken by content creators when faced with such distribution mechanisms as the Internet. As an example, for audio clips commercially exchanged on the World Wide Web (WWW), a part of the Internet, 8 bit sampled audio or audio downsampled from 44.1 kHz (CD-quality), to 22 kHz and lower. Digital filters, however, are not ideal because of trade-offs between attenuation and time-domain response, but provide the engineer or similarly-trained individual with a set of decisions to make about maximizing content quality with minimum data overhead and consideration of the ultimate delivery mechanism for the content (CDs, cable television, satellite, audio tape, stereo amplifier, etc.).
For audio signals, and, more generally, for other frequency-based content, such as video, one method of using digital filters is to include the use of an input filter to prevent frequency aliasing higher than the so-called Nyquist frequencies. The Nyquist theorem specifies that the sampling frequency must be at least twice the highest signal frequency of the sampled information (e.g., for the case of audio, human perception of audio frequencies is in a range between 20 Hz and 20 kHz, such signals should be sampled at a frequency of at least 40 KHz). Without an input filter, aliases can still occur leaving an aliased signal in the original bandwidth that cannot be removed.
Even with anti-aliasing filters, quantization error can still cause low level aliasing which may be removed with a dither technique. Dither is a method of adding random noise to the signal, and is used to de-correlate quantization error from the signal while reducing the audibility of the remaining noise. Distortion may be removed, but at the cost of adding more noise to the filtered output signal. An important effect is the subsequent randomization of the quantization error while still leaving an envelope of an unremovable signaling band of noise. Thus, dither is done at low signal levels, affecting only the least significant bits of the samples. Conversely, digital watermarks, which may take the form of essentially randomly-mapped noise, are intended to be inserted into samples of digitized content in a manner such as to maximize encoding levels while minimizing any perceivable artifacts that would indicate their presence or allow for removal by filters, and without destroying the content signal. Further, digital watermarks should be inserted with processes that necessitate random searching in the content signal for watermarks if an attacker lacks the keys. Attempts to over-encode noise into known watermarked signal locations to eliminate the information signal can be made difficult or impossible without damaging the content signal by relying on temporal encoding and randomization in the generation of keys during digital watermark insertion. As a result, although the watermark occupies only a small percentage of the signal, an attacker is forced to over-encode the entire signal at the highest encoding level, which creates audible artifacts.
The present invention relates to methods for obtaining more optimal models to design watermark systems that are tamper-resistant given the number and breadth of existent digitized sample options with differing frequency and time components (audio, video, pictures, multimedia, virtual reality, etc.).
To accomplish these goals, the present invention maintains the highest quality of a given content signal as it was mastered, with its watermarks suitably hidden, taking into account usage of digital filters and error correction presently concerned solely with the quality of content signals.
Additionally, where a watermark location is determined in a random or pseudo-random operation dependent on the creation of a pseudo-random key, as described in U.S. Pat. No. 5,613,004, and unlike other forms of manipulating digitized sample streams to improve quality or encode known frequency ranges, an engineer seeking to provide high levels of protection of copyrights, ownership, etc. is concerned with the size of a given key, the size of the watermark message and the most suitable area and method of insertion. Robustness is improved through highly redundant error correction codes and interleaving, including codes known generally as q-ary Bose-Chaudhuri-Hocquenghem (BCH) codes, a subset of Hamming coding operations, and codes combining error correction and interleaving, such as the Cross-interleave Reed-Solomon Code. Using such codes to store watermark information in the signal increases the number of changes required to obliterate a given watermark. Preprocessing the watermark by considering error correction and the introduction of random data to make watermark discovery more difficult, prior to watermarking, will help determine sufficient key size. More generally, absolute key size can be determined through preprocessing the message and the actual digital watermark (a file including information regarding the copyright owner, publisher, or some other party in the chain of exchange of the content) to compute the absolute encoded bit stream and limiting or adjusting the key size parameter to optimize the usage of key bits. The number of bits in the primary key should match or exceed the number of bits in the watermark message, to prevent redundant usage of key bits. Optimally, the number of bits in the primary key should exactly match the watermark size, since any extra bits are wasted computation.
Insertion of informational signals into content signals have been contemplated. More detailed discussions are included in related patents entitled “Steganographic Method and Device” and “Method for Human Assisted Random Key Generation and Application for Digital Watermark System.” The following discussion illustrates some previously disclosed systems and their weaknesses.
Typically, previously disclosed systems lack emphasis or implementation of any pseudo-random operations to determine the insertion location, or map, of information signals relating to the watermarks. Instead, previous implementations provide “copy protect” flags in obvious, apparent and easily removable locations. Further, previous implementations do not emphasize the alteration of the content signal upon removal of the copy protection.
Standards for digital audio tape (DAT) prescribe insertion of data, such as ISRC (Industry Standard Recording Codes) codes, title, and time in sub-code according to the Serial Copy Management System (SCMS) to prevent multiple copying of the content. One time copying is permitted, however, and systems with AES3 connectors, which essentially override copy protection in the sub-code as implemented by SCMS, actually have no copy limitations. The present invention provides improvement over this implementation with regard to the ability of unscrupulous users to load digital data into unprotected systems, such as general computing devices, that may store the audio clip in a generalized file format to be distributed over an on-line system for further duplication. The security of SCMS (Serial Copy Management System) can only exist as far as the support of similarly oriented hardware and the lack of attempts by those skilled in the art to simply remove the subcode data in question.
Previous methods seek to protect content, but the shortcomings are apparent. U.S. Pat. No. 5,319,735 to Preuss et al. discusses a spread spectrum method that would allow for over-encoding of the described, thus known, frequency range, and is severely limited in the amount of data that can be encoded—4.3 8-bit symbols per second. However, with the Preuss et al. method, randomization attacks will not result in audible artifacts in the carrier signal, or degradation of the content as the information signal is in the subaudible range. It is important to note the difference in application between spread spectrum in military field use for protection of real-time radio signals, and encoding information into static audio files. In the protection of real-time communications, spread spectrum has anti-jam features, since information is sent over several channels at once. Therefore, in order to jam the signal, one has to jam all channels, including their own. In a static audio file, however, an attacker has practically unlimited time and processing power to randomize each sub-channel in the signaling band without penalty to themselves, so the anti-jam advantages of spread spectrum do not extend to this domain.
In a completely different implementation, U.S. Pat. No. 5,379,345 to Greenberg seeks enforcement of broadcast contracts using a spread spectrum modulator to insert signals that are then confirmed by a spread spectrum-capable receiver to establish the timing and length that a given, marked advertisement is played. This information is measured against a specific master of the underlying broadcast material. The Greenberg patent does not ensure that real-time downloads of copyrighted content can be marked with identification information unless all download access points (PCs, modems, etc.), and upload points for that matter, have spread spectrum devices for monitoring.
Other methods include techniques similar to those disclosed in related patents and patent applications cited above, but lack the pseudo-random dimension of those patent applications for securing the location of the signals inserted into the content. One implementation conducted by Michael Gerzon and Peter Craven, and described by Ken Pohlmann in the Third edition of Principles of Digital Audio, illustrates a technology called “buried data technique,” but does not address the importance of randomness in establishing the insertion locations of the informational signals in a given content signal, as no pseudo-random methods are used as a basis for insertion. The overriding concern of the “buried data techniques” appears to be to provide for a “known channel” to be inserted in such a manner as to leave little or no perceivable artifacts in the content signal while prescribing the exact location of the information (i.e., replacing the least significant bits (LSB) in a given information signal). In Gerzon and Craven's example, a 20-bit signal gives way to 4-bits of LSBs for adding about 27 dB of noise to the music. Per channel data insertion reached 176.4 kilobits per second per channel, or 352.8 kbps with stereo channels. Similarly attempted data insertion by the present inventors using random data insertion yielded similar rates. The described techniques may be invaluable to manufacturers seeking to support improvements in audio, video and multimedia quality improvements. These include multiple audio channel support, surround sound, compressed information on dynamic range, or any combination of these and similar data to improve quality. Unfortunately, this does little or nothing to protect the interests of copyright holders from unscrupulous pirates, as they attempt to create unmarked, perfect copies of copyrighted works.
The present invention also relates to the “Steganographic Method and Device” patent and the “Method for Human-Assisted Random Key Generation and Application for Digital Watermark System” patent, as well as U.S. Pat. No. 5,745,569 entitled “Method for Stega-Cipher Protection of Computer Code” as mentioned above, specifically addressing the weakness of inserting informational signals or digital watermarks into known locations or known frequency ranges, which are sub-audible. The present invention seeks to improve on the methods disclosed in these patent applications and other methods by describing specific optimization techniques at the disposal of those skilled in the art. These techniques provide an a la carte method for rethinking error correction, interleaving, digital and analog filters, noise shaping, nonlinear random location mapping in digitized samples, hashing, or making unique individual watermarks, localized noise signal mimic encoding to defeat noise filtering over the entire sample stream, super audible spread spectrum techniques, watermark inversion, preanalyzing watermark key noise signatures, and derivative analysis of suspect samples against original masters to evaluate the existence of watermarks with statistical techniques.
The goal of a digital watermark system is to insert a given information signal or signals in such a manner as to leave few or no artifacts in the underlying content signal, while maximizing its encoding level and location sensitivity in the signal to force damage to the content signal when removal is attempted. The present invention establishes methods for estimating and utilizing parameters, given principles of the digitization of multimedia content (audio, video, virtual reality, etc.), to create an optimized “envelope” for insertion of watermarks, and thus establish secured responsibility for digitally sampled content. The pseudo-random key that is generated is the only map to access the information signal while not compromising the quality of the content. A digital watermark naturally resists attempts at removal because it exists as purely random or pseudo-random noise in a given digitized sample. At the same time, inversion techniques and mimicking operations, as well as encoding signal features instead of given samples, can make the removal of each and every unique encoded watermark in a given content signal economically infeasible (given the potential commercial returns of the life of a given copyright) or impossible without significantly degrading the quality of the underlying, “protected” signal. Lacking this aesthetic quality, the marketability or commercial value of the copy is correspondingly reduced.
The present invention preserves quality of underlying content signals, while using methods for quantifying this quality to identify and highlight advantageous locations for the insertion of digital watermarks.
The present invention integrates the watermark, an information signal, as closely as possible to the content signal, at a maximal level, to force degradation of the content signal when attempts are made to remove the watermarks.
General methods for watermarking digitized content, as well as computer code, are described in related patents entitled “Steganographic Method and Device” and entitled “Method for Stega-Cipher Protection of Computer Code.” Recognizing the importance of perceptual encoding of watermarks by the authors and engineers who actually create content is addressed in patent “Method for Human Assisted Random Key Generation and Application for Digital Watermark System.”
The present invention describes methods of random noise creation given the necessary consequence of improving signal quality with digitization techniques. Additionally, methods are described for optimizing projections of data redundancy and overhead in error correction methods to better define and generate parameters by which a watermarking system can successfully create random keys and watermark messages that subsequently cannot be located and erased without possession of the key that acts as the map for finding each encoded watermark. This description will provide the backdrop for establishing truly optimized watermark insertion including: use of nonlinear (chaotic) generators; error correction and data redundancy analysis to establish a system for optimizing key and watermark message length; and more general issues regarding desired quality relating to the importance of subjecting watermarked content to different models when the content may be distributed or sold in a number of prerecorded media formats or transmitted via different electronic transmission systems; this includes the use of perceptual coding; particularized methods such as noise shaping; evaluating watermark noise signatures for predictability; localized noise function mimic encoding; encoding signal features; randomizing time to sample encoding of watermarks; and, finally, a statistical method for analyzing composite watermarked content against a master sample content to allow watermark recovery. All of these features can be incorporated into specialized digital signal processing microprocessors to apply watermarks to nongeneralized computing devices, such as set-top boxes, video recorders that require time stamping or authentication, digital versatile disc (DVD) machines and a multitude of other mechanisms that play or record copyrighted content.
As discussed above, the Nyquist Theorem proves that bandlimited signals can be sampled, stored, processed, transmitted, reconstructed, desampled or processed as discrete values. In order for the theorem to hold true, the sampling must be done at a frequency that is at least twice the frequency of the highest signal frequency to be captured and reproduced. Aliasing will occur as a form of signal fold over, if the signal contains components above the Nyquist frequency. To establish the highest possible quality in a digital signal, aliasing is prevented by low-pass filtering the input signal to a given digitization system by a low-pass or anti-aliasing filter. Any residue aliasing which may result in signal distortion, relates to another area of signal quality control, namely, quantization error removal.
Quantization is required in a digitization system. Because of the continuous nature of an analog signal (amplitude vs. time), a quantized sample of the signal is an imperfect estimate of the signal sample used to encode it as a series of discrete integers. These numbers are merely estimates of the true value of the signal amplitude. The difference between the true analog value at a discrete time and the quantization value is the quantization error. The more bits allowed per sample, the greater the accuracy of estimation; however, errors still always will occur. It is the recurrent nature of quantization errors that provides an analogy with the location of digital watermarks.
Thus, methods for removal of quantization errors have relevance in methods for determining the most secure locations for placement of watermarks to prevent the removal of such watermarks.
The highest fidelity in digital reproduction of a signal occurs at points where the analog signal converges with a given quantization interval. Where there is no such convergence, in varying degrees, the quantization error will be represented by the following range:
+Q/2 and −Q/2, where Q is the quantization interval. Indeed, describing maximization of the quantization error and its ratio with the maximum signal amplitude, as measured, will yield a signal-to-error ratio (S/E) which is closely related to the analog signal-to-noise ratio (S/N). To establish more precise boundaries for determining the S/E, with root mean square (rms) quantization error E.sub.rms, and assuming a uniform probability density function 1/Q (amplitude), the following describes the error:
Signal to quantization error is expressed as:
Finally, in decibels (dB) and comparing 16-bit and 15-bit quantization:
S/E(db)=10 log [(3/2).sup.1/2(2.sup.n)].sup.2=10 log(3/2)+2.sup.n log 2
(or “=20 log [(3/2).sup.1/2(2.sup.n)]”)
This explains the S/E ratio of 98 dB for 16-bit and 92 dB for 15-bit quantization. The 1.76 factor is established statistically as a result of peak-to-rms ratio of a sinusoidal waveform, but the factor will differ if the signal waveform differs. In complex audio signals, any distortion will exist as white noise across the audible range. Low amplitude signals may alternatively suffer from distortion.
Quantization distortion is directly related with the original signal and is thus contained in the output signal. This being the case, implementation of so-called quality control of the signal may use dither. As discussed above, dither is a method of adding random noise to the signal to de-correlate quantization error from the signal while reducing the audibility of the remaining noise. Distortion may be removed at the cost of adding more noise to the filtered output signal. An important effect is the subsequent randomization of the quantization error while still leaving an envelope of an unremovable signaling band of noise. Dither, done at low signal levels, typically affects only the least significant bits of the samples.
Use of linear and nonlinear quantization can affect the output signal, and this trade-off must be considered for a system of watermarks designed to determine acceptable quantization distortion to contain the digital watermark. For audio systems, block linear quantization implementations may be chosen. However, block floating point and floating point systems, non-uniform companding, adaptive delta modulation, adaptive differential pulse-code modulation, and perceptual coding schemes (which are oriented around the design of filters that closely match the actual perception of humans) appear to provide alternative method implementations that would cause higher perceptible noise artifacts if filtering for watermarks was undertaken by pirates. The choice of method is related to the information overhead desired.
According to one aspect of the present invention, the envelope described in the quantization equations above is suitable for preanalysis of a digitized sample to evaluate optimal locations for watermarks. The present example is for audio, but corresponding applications for digitization of video may be implemented using the quantization of color and luminance.
The matter of dither complicates preanalysis of a sample evaluated for digital watermarks. Therefore, the present invention also defines the optimal envelope more closely given three types of dither (this example is for audio, others exist for video): triangular probability density function (pdf), Gaussian pdf, and rectangular pdf. Again, the purpose is to establish better boundaries for the random or pseudo-random insertion of a watermark in a region of a content signal that would represent an area for hiding watermarks in a manner most likely to cause damage to the content signal if unauthorized searches or removal are undertaken. Dither makes removal of quantization error more economical through lower data overhead in a system by shifting the signal to decorrelate errors from the underlying signal. When dither is used, the dither noise and signal are quantized together to randomize the error. Dither which is subtractive requires removing the dither signal after requantization and creates total error statistical independence. Subtractive dither also provides further parameters for digital watermark insertion given the ultimate removal of the dither signal before finalizing the production of the content signal. With nonsubtractive dither, the dither signal is permanently left in the content signal. Errors would not be independent between samples. For this reason, further analysis with the three types of dither should reveal an acceptable dither signal without materially affecting the signal quality.
Some proposed systems for implementing copyright protection into digitally-sampled content predicate the natural occurrence of artifacts that cannot be removed. Methods for creating a digital signature in the minimized error that is evident, as demonstrated by explanations of dither, point out another significant improvement over the art in the system described in the present invention and its antecedents. Every attempt is made to raise the error level of error from LSBs to a level at which erasure necessarily leads to the degradation of the “protected” content signal. Furthermore, with such a system, pirates are forced to make guesses, and then changes, at a high enough encoding level over a maximum amount of the content signal so as to cause signal degradation, because guessing naturally introduces error. Thus, dither affects the present invention's envelope by establishing a minimum encoding level. Any encoding done below the dither level might be erased by the dither.
One embodiment of the present invention may be viewed as the provision of a random-super-level non-subtractive dither which contains information (the digital watermark).
To facilitate understanding how this does not cause audible artifacts, consider the meaning of such encoding in terms of the S/E ratio. In a normal 16-bit signal, there is a 98 dB S/E according to the equation S/E=6.02n+1.76. Consider that the encoding of watermark information looks like any other error, except it moves beyond the quantization level, out of the LSBs. If the error is of a magnitude expressed in bits, for example, 8 bits, then at that moment, the signal effectively drops to 8 bits (16-8). This corresponds to a momentary drop in S/E, referred to herein as the momentary S/E. Yet, these errors are relatively few and far between and therefore, since the signal is otherwise comprised of higher-bit samples, a “Perceived S/E” may be derived which is simply the weighted average of the samples using the “Pure S/E” (the samples without watermark information) and those with the Momentary S/E. As a direct consequence, it may be observed that as the watermark map becomes more sparse, the number of errors introduced in a given range is reduced, and the higher the perceived S/E. It also helps that the error is random, and so over time, appears as white noise, which is relatively unobtrusive. In general, it is observed that as long as introduced errors leave resulting samples within an envelope in the sample window described by minimum and maximum values, before error introduction, and the map is sufficiently sparse, the effects are not perceived.
In addition, it is possible to obtain an even higher Perceived S/E by allowing the range of introduced errors to vary between a minimum and maximum amount. This makes the weighted average S/E higher by reducing the average introduced error level. Yet, someone trying to erase a watermark, assuming they knew the maximum level, would have to erase at that level throughout the data, since they would not know how the introduced level varies randomly, and would want to erase all watermarks.
A watermarking cipher could perform this operation and may also introduce the further step of local dither (or other noise) significantly above the quantization amplitude on a window by window basis, randomly, to restrict total correlation between the watermark signal and the probability that it remains independent between samples, similar to the use of subtractive dither implementations that are mostly concerned with the ultimate removal of the dither signal with requantization. This ability could be used to accomplish signal doping, which adds a degree of random errors that do not contain watermark information so as to prevent differential analysis of multiple watermarked copies. Alternatively, it could be used to mimic a specific noise function in a segment of the signal in order to defeat attempts to filter a particular type of noise over the entire signal. By varying this function between watermarks, it may help ensure that any particular filter is of little use over the whole signal. By applying several filters in series, it seems intuitive that the net results would be significantly different from the original signal.
The discussion may be more appropriately introduced with perceptual coding techniques, but a watermarking system could also defeat some detection and correction with dither by inserting watermarks into signal features, instead of signal samples. This would be equivalent to looking for signal characteristics, independent of the overall sample as it exists as a composite of a number of signals. Basically, instead of encoding on a bit per sample basis, one might spread bits over several samples. The point of doing this is that filtering and convolution operations, like “flanging,” which definitely change individual samples on a large scale, might leave intact enough of a recognizable overall signal structure (the relationship between multiple samples) to preserve the watermark information. This may be done by measuring, generalizing, and altering features determined by the relationships between samples or frequency bands. Because quantization is strictly an art of approximation, signal-to-error ratios, and thus the dynamic range of a given system are determined.
The choice of eliminating quantization distortion at the expense of leaving artifacts (not perceptible) is a permanent trade-off evident in all digitization systems which are necessarily based on approximation (the design goal of the present invention in preanalyzing a signal to mask the digital watermarks make imperceptibility possible). The high fidelity of duplication and thus subsequent ability to digitally or electronically transmit the finished content (signal) is favored by consumers and artists alike. Moreover, where there continues to be a question of approximating in quantization—digital watermark systems will have a natural partner in seeking optimized envelopes in the multitude and variety of created digitized content.
Another aspect of optimizing the insertion of digital watermarks regards error correction. Highly redundant error codes and interleaving might create a buffer against burst errors introduced into digital watermarks through randomization attacks. A detailed description follows from the nature of a digitization system—binary data can be corrected or concealed when errors exist. Random bit errors and burst errors differ in their occurrence: Random bit errors are error bits occurring in a random manner, whereas burst errors may exist over large sequences of the binary data comprising a digitized signal. Outside the scope of the present invention are errors caused by physical objects, such as dust and fingerprints, that contribute to the creation of dropouts are different from the errors addressed herein.
Measuring error with bit-error ratio (BER), block error ratio (BLER) and burst-error length (BERL), however, provides the basis of error correction. Redundancy of data is a focus of the present invention. This data necessarily relies on existing data, the underlying content. To efficiently describe optimal parameters for generating a cryptographic key and the digital watermark message discussion of error correction and error concealment techniques is important.
Forms of error detection include one-bit parity, relying on the mathematical ability to cast out numbers, for binary systems including digitization systems, such as 2. Remainders given odd or even results (parity) that are probabilistically determined to be errors in the data. For more appropriate error detection algorithms, such as Cyclic Redundancy Check Code (CRCC), which are suited for the detection of commonly occurring burst error. Pohlmann (Principles of Digital Audio) notes the high accuracy of CRCC (99.99%) and the truth of the following statements given a k-bit data word with m bits of CRCC, a code word of n bits is formed (m=n−k):
burst errors less than or equal to m bits are always predictable.
the detection probability of burst errors of m+1 bits=1−2.sup.−m.
the detection probability of burst errors longer than m+1 bits=1−2.sup.−m
random errors up to 3 consecutive bits long can be detected.
The medium of content delivery, however, provides the ultimate floor for CRCC design and the remainder of the error correction system.
Error correction techniques can be broken into three categories: methods for algebraic block codes, probabilistic methods for convolutional codes, and cross-interleave code where block codes are used in a convolution structure. As previously discussed, the general class of codes that assist in pointing out the location of error are known generally as Hamming codes, versus CRCC which is a linear block code.
What is important for establishing parameters for determining optimized error coding in systems such as digital audio are more specifically known as Reed-Solomon Codes which are effective methods for correcting burst errors. Certain embodiments of the present invention presuppose the necessity of highly redundant error codes and interleaving, such as that done in Cross Interleave Reed-Solomon Code, to counter burst errors typically resulting from randomization attacks. More generally, certain embodiments of the present invention include the use of Hamming Codes of (n, n) to provide n−1 bit error detection and n−2 bit error correction. Further, a Hamming distance of n (or greater than n) is significant because of the nature of randomization attacks. Such an attack seeks to randomize the bits of the watermark message. A bit can be either 0 or 1, so any random change has a 50% chance of actually changing a bit from what it was (50% is indicative of perfect randomness). Therefore, one must assume that a good attack will change approximately half the bits (50%). A Hamming distance of n or greater, affords redundancy on a close par with such randomization. In other words, even if half the bits are changed, it would still be possible to recover the message.
Because interleaving and parity make data robust for error avoidance, certain embodiments of the present invention seek to perform time interleaving to randomly boost momentary S/E ratio and help prevent removing keys and watermarks that may be subsequently determined not to be “errors.”
Given a particular digital content signal, parity, interleaving, delay, and cross-interleaving, used for error correction, should be taken into account when preprocessing information to compute absolute size requirements of the encoded bit stream and limiting or adjusting key size parameters to optimize and perhaps further randomize usage of key bits. In addition, these techniques minimize the impact of errors and are thus valuable in creating robust watermarks.
Uncorrected errors can be concealed in digital systems. Concealment offers a different dynamic to establish insertion parameters for the present invention. Error concealment techniques exist because it is generally more economical to hide some errors instead of requiring overly expensive encoders and decoders and huge information overheads in digitization systems. Muting, interpolation, and methods for signal restoration (removal of noise) relate to methods suggested by the present invention to invert some percentage or number of watermarks so as to ensure that at least some or as many as half of the watermarks must still remain in the content signal to effectively eliminate the other half. To invert a watermark relative to another watermark is to inverse the mathematical or logical relationships between the two watermarks (for example, without limitation, by bit flipping, by inverting the phase relationships, or by using an inverse filter relationship). Given that a recording contains noise, whether due to watermarks or not, a restoration which “removes” such noise is likely to result in the changing of some bit of the watermark message. Therefore, by inverting every other watermark, it is possible to insure that the very act of such corrections inverts enough watermark bits to create an inverse watermark. This inversion presupposes that the optimized watermark insertion is not truly optimal, given the will of a determined pirate to remove watermarks from particularly valuable content.
Ultimately, the inability to resell or openly trade unwatermarked content will help enforce, as well as dictate, the necessity of watermarked content for legal transactions.
The mechanisms discussed above reach physical limits as the intent of signal filtering and error correction are ultimately determined to be effective by humans—decidedly analog creatures. All output devices are thus also analog for playback.
The present invention allows for a preprocessed and preanalyzed signal stream and watermark data to be computed to describe an optimized envelope for the insertion of digital watermarks and creation of a pseudorandom key, for a given digitized sample stream. Randomizing the time variable in evaluating discrete sample frames of the content signal to introduce another aspect of randomization could further the successful insertion of a watermark. More importantly, aspects of perceptual coding are suitable for methods of digital watermarks or super-audible spread spectrum techniques that improve on the art described by the Preuss et al. patent described above.
The basis for a perceptual coding system, for audio, is psychoacoustics—the analysis of what the human ear is able to perceive. Similar analysis is conducted for video systems, and some may argue abused, with such approaches as “subliminal seduction” in advertising campaigns. Using the human for design goals is vastly different than describing mathematical or theoretical parameters for watermarks. On some level of digital watermark technology, the two approaches may actually complement each other and provide for a truly optimized model.
The following example applies to audio applications. However, this example and other examples provided herein are relevant to video systems as well as audio systems. Where a human ear can discern between energy inside and outside the “critical band,” (described by Harvey Fletcher) masking can be achieved. This is particularly important as quantization noise can be made imperceptible with perceptual coders given the maintenance of a sampling frequency, and decreased word length (data) based on signaling conditions. This contrasts with the necessary decrease of 6 dB/bit with decreases in the sampling frequency as described above in the explanation of the Nyquist Theorem. Indeed, data quantity can be reduced by 75%. This is an extremely important variable to feed into the preprocessor that evaluates the signal in advance of “imprinting” the digital watermark.
In multichannel systems, such as MPEG-1, AC-3 and other compression schemes, the data requirement (bits) is proportional to the square root of the number of channels. What is accomplished is masking that is nonexistent perceptually, only acoustically. The phrase “nonexistent perceptually” means merely that the masking is not perceived as being present.
Taken to another level for digital watermarking, which is necessary for content that may be compressed and decompressed, forward adaptive allocation of bits and backward adaptive allocation provide for encoding signals into content signals in such a manner that information can be conveyed in the transmission of a given content signal that is subsequently decoded to convey the relatively same audible signal to a signal that carries all of its bits—e.g., no perceptual differences between two signals that differ in bit size. This coding technique must also be preanalyzed to determine the most likely sample bits, or signal components, that will exist in the smaller sized signal. This is also clearly a means to remove digital watermarks placed into LSBs, especially when they do not contribute significant perceptible value to the analyzed signal. Further methods for data reduction coding are similarly important for preanalyzing a given content signal prior to watermarking. Frequency domain coders, such as sub-band and transform bands, can achieve data reduction of ratios between 4:1 and 12:1. The coders adaptively quantize samples in each sub-band based on the masking threshold in that sub-band (See Pohlmann, Principles of Digital Audio). Transform coders, however, convert time domain samples into the frequency domain for accomplishing lossless compression. Hybrid coders combine both sub-band and transform coding, again with the ultimate goal of reducing the overall amount of data in a given content signal without loss of perceptible quality.
With digital watermarks, descriptive analysis of an information signal is important to preanalyze a given watermark's noise signature. Analysis of this signature versus the preanalysis of the target content signal for optimized insertion location and key/message length, are potentially important components to the overall implementation of a secure watermark. It is important that the noise signature of a digital watermark be unpredictable without the pseudo-random key used to encode it. Noise shaping, thus, has important applications in the implementation of the present invention. In fact, adaptive dither signals can be designed to correlate with a signal so as to mask the additional noise—in this case a digital watermark. This relates to the above discussion of buried data techniques and becomes independently important for digital watermark systems. Each instance of a watermark, where many are added to a given content signal given the size of the content and the size of the watermark message, can be “noise shaped” and the binary description of the watermark signature may be made unique by “hashing” the data that comprises the watermark. Generally, hashing the watermark prior to insertion is recommended to establish differences between the data in each and every watermark “file.”
Additionally, the present invention provides a framework in which to analyze a composite content signal that is suspected to contain a watermarked sample of a copyrighted work, against an unwatermarked original master of the same sample to determine if the composite content actually contains a copy of a previously watermarked content signal. Such an analysis may be accomplished in the following scenario:
Assume the composite signal contains a watermark from the sample.
Assume the provision of the suspect composite signal C.sub.w(t) (w subscript denotes a possible watermark) and the unwatermarked original sample S.sub.uw(t). These are the only two recordings the analyzer is likely to have access to.
Now, it is necessary to recover a watermarked sample S.sub.w(t).
The methods of digital signal processing allow for the computation of an optimal estimate of a signal. The signal to be estimated is the composite minus the watermarked sample, or C″.sub.w(t)=C.sub.w(t)−S.sub.w−(t). The analyzer, however, cannot determine a value of S.sub.w(t), since it does not know which of the many possible S.sub.w(t) signals was used in the composite. However, a close estimate may be obtained by using S.sub.uw(t), since watermarking makes relatively minor changes to a signal.
So, C″.sub.w(t) (an estimate of C′.sub.w(t) given C″.sub.w(t) and S′.sub.w(t)) may be obtained. Once C″.sub.w(t) is calculated, it is simply subtracted from C.sub.w(t). This yields S′.sub.w(t) C.sub.w(t)−C″.sub.w(t). If the watermark is robust enough, and the estimate good enough, then S′.sub.w(t), which is approximately equal to S.sub.w(t), can be processed to extract the watermark. It is simply a matter of attempting watermark decoding against a set of likely encoding key candidates.
Note that although a watermark is initially suspected to be present in the composite, and the process as if it is, the specifics of the watermark are not known, and a watermark is never introduced into the calculations, so a watermark is extracted, it is valid, since it was not introduced by the signal processing operations.
The usefulness of this type of operation is demonstrated in the following scenario:
People are interested in simply proving that their copyrighted sample was dubbed into another recording, not the specifics of ownership of the sample used in the dubbing. So, this implies that only a single, or limited number of watermark keys would be used to mark samples, and hence, the decode key candidates are limited, since the same key would be used to encode simple copyright information which never varies from copy to copy.
There are some problems to solve to accomplish this sort of processing. The sample in question is generally of shorter duration than the composite, and its amplitude may be different from the original. Analysis techniques could use a combination of human-assisted alignment in the time domain, where graphical frequency analysis can indicate the temporal location of a signal which closely matches that of the original sample. In addition, automatic time warping algorithms which time align separate signals, on the assumption they are similar could also be used to solve temporal problems. Finally, once temporal alignment is accomplished, automatic amplitude adjustment could be performed on the original sample to provide an optimal match between the composite section containing the sample and the original sample.
It may be desirable to dynamically vary the encoding/decoding algorithm during the course of encoding/decoding a signal stream with a given watermark. There are two reasons for dynamically varying the encoding/decoding algorithm.
The first reason for dynamically varying the encoding/decoding algorithm is that the characteristics of the signal stream may change between one locality in the stream and another locality in the stream in a way that significantly changes the effects that a given encoding algorithm may have on the perception of that section of the stream on playback. In other words, one may want the encoding algorithm, and by implication, the decoding algorithm, to adapt to changes in the signal stream characteristics that cause relative changes in the effects of the encoding algorithm, so that the encoding process as a whole causes fewer artifacts, while maintaining a certain level of security or encoding a given amount of information.
The second reason for dynamically varying the encoding/decoding algorithm is simply to make more difficult attempts at decoding watermarks without keys. It is obviously a more difficult job to attempt such attacks if the encoding algorithm has been varied. This would require the attacker to guess the correct order in which to use various decoding algorithms.
In addition, other reasons for varying the encoding/decoding algorithms may arise in the future.
Two methods for varying of the encoding/decoding algorithms according to embodiments of the present invention are described herein. The first method corresponded to adaptation to changing signal characteristics. This method requires a continuous analysis of the sample windows comprising the signal stream as passed to the framework. Based on these characteristics, which are mathematically well-defined functions of the sample stream (such as RMS energy, RMS/peak ratio, RMS difference between samples—which could reflect a measure of distortion), a new CODEC (encoder/decoder) module, from among a list of pre-defined CODECs, and the algorithms implemented in them, can be applied to the window in question. For the purpose of this discussion, windows are assumed to be equivalent to frames. And, in a frame-based system, this is a straightforward application of the architecture to provide automated variance of algorithms to encode and decode a single watermark. The signal features (or signal characteristics) can also be identified from relationships between multiple sample windows in the digital content signal. For example, an implementation using a sample window size of 15 seconds can be compared to an implementation using a sample window size of 45 seconds. The comparison will reveal differences in robustness of the encoding and decoding operations—which differences will be useful in identifying signal features that are desirable to target.
The second method for varying of the encoding/decoding algorithms corresponds to increased security. This method is easier, since it does not require the relatively computationally-expensive process of further analyzing the samples in a frame passed to the Framework. In this method, the Framework selects a new CODEC, from among a list of predefined CODECs, to which to pass the sample frame as a function of the pseudo-random key employed to encode/decode the watermark. Again, this is a straightforward application of framework architecture which provides automated variance of algorithms to encode and decode a single watermark versus limitations evident in the analysis of a single random noise signal inserted over the entire content signal as proposed by Digimarc, NEC, Thorn EMI and IBM under the general guise of spread spectrum, embedded signaling schemes.
It is important to note that the modular framework architecture, in which various modules including CODECs are linked to keys, provides a basic method by which the user can manually accomplish such algorithmic variations for independent watermarks. The main difference detailed above is that an automated method to accomplish this can be used within single watermarks.
Automated analysis of composited copyrighted material offers obvious advantages over subjective “human listening” and “human viewing” methods currently used in copyright infringement cases pursued in the courts.
In addition to the embodiments discussed above, in the same manner that the signal to be watermarked, or scrambled, may be separated in time (from the beginning to the end of the signal) into streams, as disclosed in the copending U.S. patent application Ser. No. 09/594,719, entitled “Utilizing Data Reduction in Steganographic and Cryptographic Systems,” (issued as U.S. Pat. No. 7,123,718 on Oct. 17, 2006) additional processing may occur prior to the actual embedding of the watermark (“preprocessing”) as well as after each instance of watermark embedding (“postprocessing”). The pre- and post-processing may be used to optimize the actual embedding process.
In cases where the intent is to watermark a compressed file, for a given signal, the coefficients that are to be manipulated may first be identified. That is, for a particular signal, such as a song, a video, etc., to be embedded in such a manner as to authorize a plurality of unique descendant copies of that signal, a preprocessing step may first complete the necessary psychoacoustic or psychovisual modeling inherent in the embedding process for that signal. In one embodiment, the watermark message may change for each descendant copy. In another embodiment, the key may change for each descendant copy to reflect the uniqueness of a given message. In either case, both the message and the key may be processed so as to reflect any intended differences among the descendant copies.
Of particular interest are those cases in which the psychoacoustic or psychovisual model may be roughly correct for any number of proprietary compression techniques (e.g., MP3, AAC, ePAC, etc. for audio, MPEG for video) and a given signal may need to be prepared so as to enable the watermarking encoder to handle multiple requests for the same signal, but for different compression schemes. The preprocessing step of the present invention provides a “skeleton” of the candidate bits, and eliminates the need to process all of the signal data each time an encoding is performed. Each time a descendant copy is created, the appropriate model (e.g., psychovisual, psychoacoustic, etc.) may be generated for that signal, or a previously saved version of the appropriate model is referenced, and then any concatenation between the model and the particular compression scheme may be matched. Any uniquely-generated descendant copy may differ from other copies in that the keys or messages may change to reflect the difference after watermark(s) have been embedded. One signal may have a plurality of watermarks, which may be a combination of different messages, different encoding keys, and different decoding keys, or key pairs, for each embedded message.
The psychoacoustic model itself may be saved for later handling in the watermark encoding process so as to decrease the requisite time to make every instance of a descendant copy unique for any number of characteristics. These characteristics may include, inter alia, a geographical territory, a transaction identification, an individual identification, a use limitation, a domain, a logical constraint, etc.
The preprocessing of the signal may prepare a key for additional postprocessing in making the key or watermarked signal unique.
Other embodiments and uses of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. The specification and examples should be considered exemplary only with the true scope and spirit of the invention indicated by the following claims. As will be easily understood by those of ordinary skill in the art, variations and modifications of each of the disclosed embodiments can be easily made within the scope of this invention as defined by the following claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3947825||May 7, 1975||Mar 30, 1976||International Business Machines Corporation||Abstracting system for index search machine|
|US3984624||Jul 25, 1974||Oct 5, 1976||Weston Instruments, Inc.||Video system for conveying digital and analog information|
|US3986624||Nov 16, 1973||Oct 19, 1976||International Incinerators, Inc.||Disposal feeding system including selective container control|
|US4038596||Jul 2, 1975||Jul 26, 1977||E M I Limited||Method and apparatus for authenticating a record medium consisting of applying two different strength magnetizing fields and monitoring the remanent fields|
|US4200770||Sep 6, 1977||Apr 29, 1980||Stanford University||Cryptographic apparatus and method|
|US4218582||Oct 6, 1977||Aug 19, 1980||The Board Of Trustees Of The Leland Stanford Junior University||Public key cryptographic apparatus and method|
|US4339134||Jul 5, 1977||Jul 13, 1982||Rockwell International Corporation||Electronic card game|
|US4390898||Mar 23, 1981||Jun 28, 1983||Northern Telecom Limited||Scrambling and unscrambling video signals in a pay TV system|
|US4405829||Dec 14, 1977||Sep 20, 1983||Massachusetts Institute Of Technology||Cryptographic communications system and method|
|US4424414||May 1, 1978||Jan 3, 1984||Board Of Trustees Of The Leland Stanford Junior University||Exponentiation cryptographic apparatus and method|
|US4528588||Sep 25, 1981||Jul 9, 1985||Loefberg Bo||Method and apparatus for marking the information content of an information carrying signal|
|US4672605||Mar 20, 1984||Jun 9, 1987||Applied Spectrum Technologies, Inc.||Data and voice communications system|
|US4748668||Jul 9, 1986||May 31, 1988||Yeda Research And Development Company Limited||Method, apparatus and article for identification and signature|
|US4789928||Jan 30, 1987||Dec 6, 1988||Flex Japan Inc.||Auction information transmission processing|
|US4827508||Oct 14, 1986||May 2, 1989||Personal Library Software, Inc.||Database usage metering and protection system and method|
|US4876617||May 5, 1987||Oct 24, 1989||Thorn Emi Plc||Signal identification|
|US4896275||Jul 10, 1987||Jan 23, 1990||Bull Hn Information Systems Inc.||Full page graphics image display data reduction|
|US4908873||May 13, 1983||Mar 13, 1990||Philibert Alex C||Document reproduction security system|
|US4939515||Sep 30, 1988||Jul 3, 1990||General Electric Company||Digital signal encoding and decoding apparatus|
|US4969204||Nov 29, 1989||Nov 6, 1990||Eastman Kodak Company||Hybrid residual-based hierarchical storage and display method for high resolution digital images in a multiuse environment|
|US4972471||May 15, 1989||Nov 20, 1990||Gary Gross||Encoding system|
|US4977594||Feb 16, 1989||Dec 11, 1990||Electronic Publishing Resources, Inc.||Database usage metering and protection system and method|
|US4979210||Jul 8, 1988||Dec 18, 1990||Matsushita Electric Industrial Co., Ltd.||Method and apparatus for protection of signal copy|
|US4980782||Jun 3, 1985||Dec 25, 1990||Peter Ginkel||Software protection and identification system|
|US5050213||Aug 6, 1990||Sep 17, 1991||Electronic Publishing Resources, Inc.||Database usage metering and protection system and method|
|US5073925||Jun 13, 1990||Dec 17, 1991||Matsushita Electric Industrial Co., Ltd.||Method and apparatus for the protection of signal copy|
|US5077665||May 25, 1989||Dec 31, 1991||Reuters Limited||Distributed matching system|
|US5113437||Oct 25, 1989||May 12, 1992||Thorn Emi Plc||Signal identification system|
|US5136581||Jul 2, 1990||Aug 4, 1992||At&T Bell Laboratories||Arrangement for reserving and allocating a plurality of competing demands for an ordered bus communication network|
|US5136646||Mar 8, 1991||Aug 4, 1992||Bell Communications Research, Inc.||Digital document time-stamping with catenate certificate|
|US5136647||Aug 2, 1990||Aug 4, 1992||Bell Communications Research, Inc.||Method for secure time-stamping of digital documents|
|US5142576||May 6, 1991||Aug 25, 1992||Market Data Corporation||System for securely providing restricted video information|
|US5161210||Nov 8, 1989||Nov 3, 1992||U.S. Philips Corporation||Coder for incorporating an auxiliary information signal in a digital audio signal, decoder for recovering such signals from the combined signal, and record carrier having such combined signal recorded thereon|
|US5210820||May 2, 1990||May 11, 1993||Broadcast Data Systems Limited Partnership||Signal recognition system and method|
|US5243423||Dec 20, 1991||Sep 7, 1993||A. C. Nielsen Company||Spread spectrum digital data transmission over TV video|
|US5243515||Oct 30, 1990||Sep 7, 1993||Lee Wayne M||Secure teleprocessing bidding system|
|US5287407||May 23, 1991||Feb 15, 1994||International Business Machines Corporation||Computer software protection|
|US5319735||Dec 17, 1991||Jun 7, 1994||Bolt Beranek And Newman Inc.||Embedded signalling|
|US5341429||Dec 4, 1992||Aug 23, 1994||Testdrive Corporation||Transformation of ephemeral material|
|US5341477||Aug 6, 1993||Aug 23, 1994||Digital Equipment Corporation||Broker for computer network server selection|
|US5363448||Jun 30, 1993||Nov 8, 1994||United Technologies Automotive, Inc.||Pseudorandom number generation and cryptographic authentication|
|US5365586||Apr 9, 1993||Nov 15, 1994||Washington University||Method and apparatus for fingerprinting magnetic media|
|US5369707||Jan 27, 1993||Nov 29, 1994||Tecsec Incorporated||Secure network method and apparatus|
|US5379345||Jan 29, 1993||Jan 3, 1995||Radio Audit Systems, Inc.||Method and apparatus for the processing of encoded data in conjunction with an audio broadcast|
|US5394324||Dec 8, 1993||Feb 28, 1995||Xerox Corporation||Auction-based control system for energy resource management in a building|
|US5398285||Dec 30, 1993||Mar 14, 1995||Motorola, Inc.||Method for generating a password using public key cryptography|
|US5406627||Dec 28, 1992||Apr 11, 1995||Nec Home Electronics, Ltd.||Digital data cryptographic system|
|US5408505||Oct 29, 1993||Apr 18, 1995||Washington University||Method and apparatus for process control, tension control, and testing of magnetic media|
|US5410598||Sep 27, 1994||Apr 25, 1995||Electronic Publishing Resources, Inc.||Database usage metering and protection system and method|
|US5412718||Sep 13, 1993||May 2, 1995||Institute Of Systems Science||Method for utilizing medium nonuniformities to minimize unauthorized duplication of digital information|
|US5428606||Jun 30, 1993||Jun 27, 1995||Moskowitz; Scott A.||Digital information commodities exchange|
|US5450490||Mar 31, 1994||Sep 12, 1995||The Arbitron Company||Apparatus and methods for including codes in audio signals and decoding|
|US5469536||May 21, 1993||Nov 21, 1995||Imageware Software, Inc.||Image editing system including masking capability|
|US5471533||May 20, 1994||Nov 28, 1995||Symbol Technologies, Inc.||Record with encoded data|
|US5478990||Oct 14, 1993||Dec 26, 1995||Coleman Environmental Systems, Inc.||Method for tracking the production history of food products|
|US5479210||Jun 7, 1994||Dec 26, 1995||Quantel, Ltd.||Video image processing system having variable data compression|
|US5487168||Dec 20, 1994||Jan 23, 1996||International Business Machines Corporation||Method and system for global optimization of device allocation|
|US5493677||Jun 8, 1994||Feb 20, 1996||Systems Research & Applications Corporation||Generation, archiving, and retrieval of digital images with evoked suggestion-set captions and natural language interface|
|US5497419||Apr 19, 1994||Mar 5, 1996||Prima Facie, Inc.||Method and apparatus for recording sensor data|
|US5506795||Feb 18, 1993||Apr 9, 1996||Yamakawa; Takeshi||Apparatus and method for generating chaotic signals and chaos device|
|US5513126||Oct 4, 1993||Apr 30, 1996||Xerox Corporation||Network having selectively accessible recipient prioritized communication channel profiles|
|US5513261||Dec 29, 1993||Apr 30, 1996||At&T Corp.||Key management scheme for use with electronic cards|
|US5530739||Mar 18, 1992||Jun 25, 1996||Fujitsu Limited||Mail center management system|
|US5530751||Jun 30, 1994||Jun 25, 1996||Hewlett-Packard Company||Embedded hidden identification codes in digital objects|
|US5530759||Feb 1, 1995||Jun 25, 1996||International Business Machines Corporation||Color correct digital watermarking of images|
|US5548579||Jun 21, 1995||Aug 20, 1996||International Business Machines Corporation||System for effective allocation of network-wide bandwidth|
|US5568570||Sep 30, 1994||Oct 22, 1996||Eastman Kodak Company||Method and apparatus for reducing quantization artifacts in a hierarchical image storage and retrieval system|
|US5579124||Feb 28, 1995||Nov 26, 1996||The Arbitron Company||Method and apparatus for encoding/decoding broadcast or recorded segments and monitoring audience exposure thereto|
|US5581703||Jun 29, 1993||Dec 3, 1996||International Business Machines Corporation||Method and apparatus for reserving system resources to assure quality of service|
|US5583488||Apr 28, 1995||Dec 10, 1996||Sala; Nicola R.||Proximity alarm system|
|US5598470||Apr 25, 1994||Jan 28, 1997||International Business Machines Corporation||Method and apparatus for enabling trial period use of software products: Method and apparatus for utilizing a decryption block|
|US5606609||Sep 19, 1994||Feb 25, 1997||Scientific-Atlanta||Electronic document verification system and method|
|US5613004||Jun 7, 1995||Mar 18, 1997||The Dice Company||Steganographic method and device|
|US5617119||Jun 7, 1995||Apr 1, 1997||Systems Research & Applications Corporation||Protection of an electronically stored image in a first color space by the alteration of a digital component in a second color space|
|US5625690||Nov 15, 1993||Apr 29, 1997||Lucent Technologies Inc.||Software pay per use system|
|US5629980||Nov 23, 1994||May 13, 1997||Xerox Corporation||System for controlling the distribution and use of digital works|
|US5633932||Dec 19, 1995||May 27, 1997||Intel Corporation||Apparatus and method for preventing disclosure through user-authentication at a printing node|
|US5634040||Dec 18, 1995||May 27, 1997||Samsung Electronics Co., Ltd.||Data communication apparatus and method having concurrent image overlay function|
|US5636276||Apr 18, 1995||Jun 3, 1997||Brugger; Rolf||Device for the distribution of music information in digital form|
|US5636292||May 8, 1995||Jun 3, 1997||Digimarc Corporation||Steganography methods employing embedded calibration data|
|US5640569||Apr 28, 1995||Jun 17, 1997||Sun Microsystems, Inc.||Diverse goods arbitration system and method for allocating resources in a distributed computer system|
|US5646997||Dec 14, 1994||Jul 8, 1997||Barton; James M.||Method and apparatus for embedding authentication information within digital data|
|US5657461||Oct 4, 1993||Aug 12, 1997||Xerox Corporation||User interface for defining and automatically transmitting data according to preferred communication channels|
|US5659726||Feb 23, 1995||Aug 19, 1997||Sandford, Ii; Maxwell T.||Data embedding|
|US5664018||Mar 12, 1996||Sep 2, 1997||Leighton; Frank Thomson||Watermarking process resilient to collusion attacks|
|US5673316||Mar 29, 1996||Sep 30, 1997||International Business Machines Corporation||Creation and distribution of cryptographic envelope|
|US5677952||Dec 6, 1994||Oct 14, 1997||International Business Machines Corporation||Method to protect information on a computer storage device|
|US5680462||Aug 7, 1995||Oct 21, 1997||Sandia Corporation||Information encoder/decoder using chaotic systems|
|US5687236||Dec 31, 1996||Nov 11, 1997||The Dice Company||Steganographic method and device|
|US5689587||Feb 9, 1996||Nov 18, 1997||Massachusetts Institute Of Technology||Method and apparatus for data hiding in images|
|US5696828||Sep 22, 1995||Dec 9, 1997||United Technologies Automotive, Inc.||Random number generating system and process based on chaos|
|US5719937||Sep 12, 1996||Feb 17, 1998||Solana Technology Develpment Corporation||Multi-media copy management system|
|US5721788||Jul 31, 1992||Feb 24, 1998||Corbis Corporation||Method and system for digital image signatures|
|US5734752||Sep 24, 1996||Mar 31, 1998||Xerox Corporation||Digital watermarking using stochastic screen patterns|
|US5737416||Aug 2, 1996||Apr 7, 1998||International Business Machines Corporation||Method and apparatus for enabling trial period use of software products: method and apparatus for utilizing a decryption stub|
|US5737733||Sep 26, 1996||Apr 7, 1998||Microsoft Corporation||Method and system for searching compressed data|
|US5740244||May 7, 1996||Apr 14, 1998||Washington University||Method and apparatus for improved fingerprinting and authenticating various magnetic media|
|US5745569||Jan 17, 1996||Apr 28, 1998||The Dice Company||Method for stega-cipher protection of computer code|
|US5748763||May 8, 1995||May 5, 1998||Digimarc Corporation||Image steganography system featuring perceptually adaptive and globally scalable signal embedding|
|US5748783 *||May 8, 1995||May 5, 1998||Digimarc Corporation||Method and apparatus for robust information coding|
|US5751811||Nov 26, 1996||May 12, 1998||Magnotti; Joseph C.||32N +D bit key encryption-decryption system using chaos|
|US5754697||May 15, 1995||May 19, 1998||Fu; Chi-Yung||Selective document image data compression technique|
|US5757923||Apr 19, 1996||May 26, 1998||Ut Automotive Dearborn, Inc.||Method of generating secret identification numbers|
|US5765152||Oct 13, 1995||Jun 9, 1998||Trustees Of Dartmouth College||System and method for managing copyrighted electronic media|
|US5768396||Jun 24, 1996||Jun 16, 1998||Yamaha Corporation||Online karaoke system with flying start performance|
|US5774452||Mar 14, 1995||Jun 30, 1998||Aris Technologies, Inc.||Apparatus and method for encoding and decoding information in audio signals|
|US5790677||Jun 29, 1995||Aug 4, 1998||Microsoft Corporation||System and method for secure electronic commerce transactions|
|US5799083||Aug 26, 1996||Aug 25, 1998||Brothers; Harlan Jay||Event verification system|
|US5809160||Nov 12, 1997||Sep 15, 1998||Digimarc Corporation||Method for encoding auxiliary data within a source signal|
|US5828325||Apr 3, 1996||Oct 27, 1998||Aris Technologies, Inc.||Apparatus and method for encoding and decoding information in analog signals|
|US5832119||Sep 25, 1995||Nov 3, 1998||Digimarc Corporation||Methods for controlling systems using control signals embedded in empirical data|
|US5850481||May 8, 1995||Dec 15, 1998||Digimarc Corporation||Steganographic system|
|US5859920||Nov 30, 1995||Jan 12, 1999||Eastman Kodak Company||Method for embedding digital information in an image|
|US5860099||May 12, 1993||Jan 12, 1999||Usar Systems, Inc.||Stored program system with protected memory and secure signature extraction|
|US5862260||May 16, 1996||Jan 19, 1999||Digimarc Corporation||Methods for surveying dissemination of proprietary empirical data|
|US5884033||May 15, 1996||Mar 16, 1999||Spyglass, Inc.||Internet filtering system for filtering data transferred over the internet utilizing immediate and deferred filtering actions|
|US5889868||Jul 2, 1996||Mar 30, 1999||The Dice Company||Optimization methods for the insertion, protection, and detection of digital watermarks in digitized data|
|US5893067||May 31, 1996||Apr 6, 1999||Massachusetts Institute Of Technology||Method and apparatus for echo data hiding in audio signals|
|US5894521||Aug 15, 1997||Apr 13, 1999||Direct Business Technologies, Inc.||System and method for encrypting sensitive information|
|US5903721||Mar 13, 1997||May 11, 1999||cha|Technologies Services, Inc.||Method and system for secure online transaction processing|
|US5905800||Mar 25, 1998||May 18, 1999||The Dice Company||Method and system for digital watermarking|
|US5905975||Jan 2, 1997||May 18, 1999||Ausubel; Lawrence M.||Computer implemented methods and apparatus for auctions|
|US5912972||Mar 26, 1997||Jun 15, 1999||Sony Corporation||Method and apparatus for embedding authentication information within digital data|
|US5917915||Jun 20, 1995||Jun 29, 1999||Sony Corporation||Scramble/descramble method and apparatus for data broadcasting|
|US5918223||Jul 21, 1997||Jun 29, 1999||Muscle Fish||Method and article of manufacture for content-based analysis, storage, retrieval, and segmentation of audio information|
|US5920900||Dec 30, 1996||Jul 6, 1999||Cabletron Systems, Inc.||Hash-based translation method and apparatus with multiple level collision resolution|
|US5923763||Mar 21, 1996||Jul 13, 1999||Walker Asset Management Limited Partnership||Method and apparatus for secure document timestamping|
|US5930369||Sep 10, 1997||Jul 27, 1999||Nec Research Institute, Inc.||Secure spread spectrum watermarking for multimedia data|
|US5930377||May 7, 1998||Jul 27, 1999||Digimarc Corporation||Method for image encoding|
|US5943422||Aug 12, 1996||Aug 24, 1999||Intertrust Technologies Corp.||Steganographic techniques for securely delivering electronic digital rights management control information over insecure communication channels|
|US5963909||Nov 25, 1997||Oct 5, 1999||Solana Technology Development Corporation||Multi-media copy management system|
|US5973731||May 30, 1995||Oct 26, 1999||Schwab; Barry H.||Secure identification system|
|US5974141||Oct 18, 1996||Oct 26, 1999||Mitsubishi Corporation||Data management system|
|US5999217||Mar 11, 1997||Dec 7, 1999||Berners-Lee; Charles Peter||Apparatus and method for encoding data|
|US6029126||Jun 30, 1998||Feb 22, 2000||Microsoft Corporation||Scalable audio coder and decoder|
|US6041316||Jul 25, 1994||Mar 21, 2000||Lucent Technologies Inc.||Method and system for ensuring royalty payments for data delivered over a network|
|US6044471||Jun 4, 1998||Mar 28, 2000||Z4 Technologies, Inc.||Method and apparatus for securing software to reduce unauthorized use|
|US6049838||Jul 1, 1996||Apr 11, 2000||Sun Microsystems, Inc.||Persistent distributed capabilities|
|US6051029||Oct 31, 1997||Apr 18, 2000||Entelos, Inc.||Method of generating a display for a dynamic simulation model utilizing node and link representations|
|US6081251||May 9, 1995||Jun 27, 2000||Sony Corporation||Apparatus and method for managing picture data|
|US6088455||Jan 7, 1997||Jul 11, 2000||Logan; James D.||Methods and apparatus for selectively reproducing segments of broadcast programming|
|US6131162||Jun 4, 1998||Oct 10, 2000||Hitachi Ltd.||Digital data authentication method|
|US6141753||Feb 10, 1998||Oct 31, 2000||Fraunhofer Gesellschaft||Secure distribution of digital representations|
|US6141754||Nov 28, 1997||Oct 31, 2000||International Business Machines Corporation||Integrated method and system for controlling information access and distribution|
|US6199058||May 17, 1999||Mar 6, 2001||Oracle Corporation||Report server caching|
|US6205249||Apr 2, 1998||Mar 20, 2001||Scott A. Moskowitz||Multiple transform utilization and applications for secure digital watermarking|
|US6208745||Dec 30, 1997||Mar 27, 2001||Sarnoff Corporation||Method and apparatus for imbedding a watermark into a bitstream representation of a digital image sequence|
|US6230268||Jul 31, 1998||May 8, 2001||International Business Machines Corporation||Data control system|
|US6233347||Dec 7, 1998||May 15, 2001||Massachusetts Institute Of Technology||System method, and product for information embedding using an ensemble of non-intersecting embedding generators|
|US6233684||Oct 10, 1997||May 15, 2001||Contenaguard Holdings, Inc.||System for controlling the distribution and use of rendered digital works through watermaking|
|US6240121||Jul 9, 1998||May 29, 2001||Matsushita Electric Industrial Co., Ltd.||Apparatus and method for watermark data insertion and apparatus and method for watermark data detection|
|US6263313||Nov 30, 1998||Jul 17, 2001||International Business Machines Corporation||Method and apparatus to create encoded digital content|
|US6272634||Aug 27, 1997||Aug 7, 2001||Regents Of The University Of Minnesota||Digital watermarking to resolve multiple claims of ownership|
|US6275988||Jun 26, 1996||Aug 14, 2001||Canon Kabushiki Kaisha||Image transmission apparatus, image transmission system, and communication apparatus|
|US6278780||Oct 29, 1998||Aug 21, 2001||Nec Corporation||Method of and an apparatus for generating internal crypto-keys|
|US6278791||May 7, 1998||Aug 21, 2001||Eastman Kodak Company||Lossless recovery of an original image containing embedded data|
|US6282300||Jan 21, 2000||Aug 28, 2001||Signafy, Inc.||Rotation, scale, and translation resilient public watermarking for images using a log-polar fourier transform|
|US6282650||Jan 25, 1999||Aug 28, 2001||Intel Corporation||Secure public digital watermark|
|US6285775||Oct 1, 1998||Sep 4, 2001||The Trustees Of The University Of Princeton||Watermarking scheme for image authentication|
|US6301663||Nov 19, 1998||Oct 9, 2001||Kabushiki Kaisha Toshiba||Copy protection apparatus and information recording medium used in this copy protection apparatus|
|US6330335||Jan 13, 2000||Dec 11, 2001||Digimarc Corporation||Audio steganography|
|US6330672||Jun 30, 1998||Dec 11, 2001||At&T Corp.||Method and apparatus for watermarking digital bitstreams|
|US6345100||Oct 14, 1998||Feb 5, 2002||Liquid Audio, Inc.||Robust watermark method and apparatus for digital signals|
|US6351765||Mar 9, 1998||Feb 26, 2002||Media 100, Inc.||Nonlinear video editing system|
|US6363483||Nov 3, 1994||Mar 26, 2002||Lucent Technologies Inc.||Methods and systems for performing article authentication|
|US6373960||Jan 6, 1998||Apr 16, 2002||Pixel Tools Corporation||Embedding watermarks into compressed video data|
|US6377625||Dec 6, 1999||Apr 23, 2002||Soft4D Co., Ltd.||Method and apparatus for generating steroscopic image using MPEG data|
|US6381618||Jun 17, 1999||Apr 30, 2002||International Business Machines Corporation||Method and apparatus for autosynchronizing distributed versions of documents|
|US6381747||Mar 31, 1997||Apr 30, 2002||Macrovision Corp.||Method for controlling copy protection in digital video networks|
|US6385329||Jul 19, 2000||May 7, 2002||Digimarc Corporation||Wavelet domain watermarks|
|US6389538||Oct 22, 1998||May 14, 2002||International Business Machines Corporation||System for tracking end-user electronic content usage|
|US6405203||Apr 21, 1999||Jun 11, 2002||Research Investment Network, Inc.||Method and program product for preventing unauthorized users from using the content of an electronic storage medium|
|US6415041||May 27, 1999||Jul 2, 2002||Nec Corporation||Digital watermark insertion system and digital watermark characteristic table creating device|
|US6425081||Aug 14, 1998||Jul 23, 2002||Canon Kabushiki Kaisha||Electronic watermark system electronic information distribution system and image filing apparatus|
|US6430301||Aug 30, 2000||Aug 6, 2002||Verance Corporation||Formation and analysis of signals with common and transaction watermarks|
|US6430302||Jan 10, 2001||Aug 6, 2002||Digimarc Corporation||Steganographically encoding a first image in accordance with a second image|
|US6442283||Jan 11, 1999||Aug 27, 2002||Digimarc Corporation||Multimedia data embedding|
|US6446211||Nov 10, 2000||Sep 3, 2002||Z4 Technologies, Inc.||Method and apparatus for monitoring software using encryption|
|US6453252||May 15, 2000||Sep 17, 2002||Creative Technology Ltd.||Process for identifying audio content|
|US6457058||Jul 12, 2001||Sep 24, 2002||Cisco Technology, Inc.||Network switch with hash table look up|
|US6463468||Jun 1, 1999||Oct 8, 2002||Netzero, Inc.||Targeted network video download interface|
|US6484264||May 18, 2000||Nov 19, 2002||Z4 Technologies, Inc.||Method for providing repeated contact with software end-user using authorized administrator|
|US6493457||Nov 16, 1998||Dec 10, 2002||At&T Corp.||Electronic watermarking in the compressed domain utilizing perceptual coding|
|US6502195||May 18, 2000||Dec 31, 2002||Z4 Technologies, Inc.||Computer readable storage medium for providing repeated contact with software end-user|
|US6522769||May 18, 2000||Feb 18, 2003||Digimarc Corporation||Reconfiguring a watermark detector|
|US6523113||Nov 9, 1998||Feb 18, 2003||Apple Computer, Inc.||Method and apparatus for copy protection|
|US6530021||Jun 2, 1999||Mar 4, 2003||Koninklijke Philips Electronics N.V.||Method and system for preventing unauthorized playback of broadcasted digital data streams|
|US6532284||Feb 27, 2001||Mar 11, 2003||Morgan Guaranty Trust Company||Method and system for optimizing bandwidth cost via caching and other network transmission delaying techniques|
|US6557103||Feb 11, 1999||Apr 29, 2003||The United States Of America As Represented By The Secretary Of The Army||Spread spectrum image steganography|
|US6584125||Dec 22, 1998||Jun 24, 2003||Nec Corporation||Coding/decoding apparatus, coding/decoding system and multiplexed bit stream|
|US6587837||Dec 1, 1998||Jul 1, 2003||International Business Machines Corporation||Method for delivering electronic content from an online store|
|US6598162||Mar 24, 1998||Jul 22, 2003||Scott A. Moskowitz||Method for combining transfer functions with predetermined key creation|
|US6606393||Dec 2, 1999||Aug 12, 2003||Verizon Laboratories Inc.||Message authentication code using image histograms|
|US6647424||May 19, 1999||Nov 11, 2003||Nortel Networks Limited||Method and apparatus for discarding data packets|
|US6658010||Jul 24, 1997||Dec 2, 2003||Hybrid Networks, Inc.||High-speed internet access system|
|US6665489||Apr 21, 1999||Dec 16, 2003||Research Investment Network, Inc.||System, method and article of manufacturing for authorizing the use of electronic content utilizing a laser-centric medium and a network server|
|US6668246||Mar 24, 1999||Dec 23, 2003||Intel Corporation||Multimedia data delivery and playback system with multi-level content and privacy protection|
|US6668325||Jun 9, 1998||Dec 23, 2003||Intertrust Technologies||Obfuscation techniques for enhancing software security|
|US6725372||Dec 2, 1999||Apr 20, 2004||Verizon Laboratories Inc.||Digital watermarking|
|US6754822||Apr 30, 1998||Jun 22, 2004||Fraunhofer-Gesellschaft Zur Forderung Der Angewandten Forshung E.V.||Active watermarks and watermark agents|
|US6775772||Oct 12, 1999||Aug 10, 2004||International Business Machines Corporation||Piggy-backed key exchange protocol for providing secure low-overhead browser connections from a client to a server using a trusted third party|
|US6784354||Mar 13, 2003||Aug 31, 2004||Microsoft Corporation||Generating a music snippet|
|US6785815||Jun 7, 2000||Aug 31, 2004||Intertrust Technologies Corp.||Methods and systems for encoding and protecting data using digital signature and watermarking techniques|
|US6785825||May 31, 2002||Aug 31, 2004||Z4 Technologies, Inc.||Method for securing software to decrease software piracy|
|US6792548||Nov 5, 2001||Sep 14, 2004||Z4 Technologies, Inc.||Method for providing repeated contact with software end-user using authorized administrator|
|US6792549||Nov 5, 2001||Sep 14, 2004||Z4 Technologies, Inc.||Method and apparatus for repeated contact of software end-user|
|US6795925||Nov 5, 2001||Sep 21, 2004||Z4 Technologies, Inc.||Computer readable storage medium for providing repeated contact with software end-user|
|US6799277||Mar 27, 2001||Sep 28, 2004||Z4 Technologies, Inc.||System and method for monitoring software|
|US6813717||May 13, 2002||Nov 2, 2004||Z4 Technologies, Inc.||Method for securing software to reduce unauthorized use|
|US6813718||Jun 26, 2002||Nov 2, 2004||Z4 Technologies, Inc.||Computer readable storage medium for securing software to reduce unauthorized use|
|US6823455||Apr 8, 1999||Nov 23, 2004||Intel Corporation||Method for robust watermarking of content|
|US6834308||Feb 17, 2000||Dec 21, 2004||Audible Magic Corporation||Method and apparatus for identifying media content presented on a media playing device|
|US6842862||Jan 10, 2003||Jan 11, 2005||Cloakware Corporation||Tamper resistant software encoding|
|US6857078||May 30, 2002||Feb 15, 2005||Z4 Technologies, Inc.||Method for securing software to increase license compliance|
|US6931534||Nov 19, 1999||Aug 16, 2005||Telefonaktiebolaget Lm Erricsson (Publ)||Method and a device for encryption of images|
|US6966002||Jun 9, 1999||Nov 15, 2005||Trymedia Systems, Inc.||Methods and apparatus for secure distribution of software|
|US6977894||May 19, 1999||Dec 20, 2005||Nortel Networks Limited||Method and apparatus for discarding data packets through the use of descriptors|
|US6978370||Mar 23, 1999||Dec 20, 2005||Cryptography Research, Inc.||Method and system for copy-prevention of digital copyright works|
|US6983337||Dec 18, 2002||Jan 3, 2006||Intel Corporation||Method, system, and program for handling device interrupts|
|US6986063||Feb 4, 2003||Jan 10, 2006||Z4 Technologies, Inc.||Method for monitoring software using encryption including digital signatures/certificates|
|US7020285||Jul 12, 2000||Mar 28, 2006||Microsoft Corporation||Stealthy audio watermarking|
|US7043050||May 2, 2001||May 9, 2006||Microsoft Corporation||Software anti-piracy systems and methods utilizing certificates with digital content|
|US7046808||Mar 24, 2000||May 16, 2006||Verance Corporation||Method and apparatus for detecting processing stages applied to a signal|
|US7050396||Nov 30, 2000||May 23, 2006||Cisco Technology, Inc.||Method and apparatus for automatically establishing bi-directional differentiated services treatment of flows in a network|
|US7051208||Oct 21, 2004||May 23, 2006||Microsoft Corporation||Technique for producing through watermarking highly tamper-resistant executable code and resulting “watermarked” code so formed|
|US7058570||Feb 10, 2000||Jun 6, 2006||Matsushita Electric Industrial Co., Ltd.||Computer-implemented method and apparatus for audio data hiding|
|US7093295||Oct 15, 1999||Aug 15, 2006||Makoto Saito||Method and device for protecting digital data by double re-encryption|
|US7103184||May 9, 2002||Sep 5, 2006||Intel Corporation||System and method for sign mask encryption and decryption|
|US7107451||Feb 22, 2001||Sep 12, 2006||Wistaria Trading, Inc.||Optimization methods for the insertion, protection, and detection of digital watermarks in digital data|
|US7150003||Nov 25, 2002||Dec 12, 2006||Matsushita Electric Industrial Co., Ltd.||Class coalescence for obfuscation of object-oriented software|
|US7162642||Jun 15, 2001||Jan 9, 2007||Digital Video Express, L.P.||Digital content distribution system and method|
|US7177430||Oct 31, 2001||Feb 13, 2007||Portalplayer, Inc.||Digital entroping for digital audio reproductions|
|US7206649||Oct 21, 2004||Apr 17, 2007||Microsoft Corporation||Audio watermarking with dual watermarks|
|US7231524||Feb 16, 2006||Jun 12, 2007||Microsoft Corporation||Method for watermarking computer programs|
|US7233669||Dec 13, 2002||Jun 19, 2007||Sony Corporation||Selective encryption to enable multiple decryption keys|
|US7240210||Nov 4, 2004||Jul 3, 2007||Microsoft Corporation||Hash value computer of content of digital signals|
|US7266697||May 3, 2004||Sep 4, 2007||Microsoft Corporation||Stealthy audio watermarking|
|US7363278||Apr 3, 2002||Apr 22, 2008||Audible Magic Corporation||Copyright detection and protection system and method|
|US7460994||Jun 20, 2002||Dec 2, 2008||M2Any Gmbh||Method and apparatus for producing a fingerprint, and method and apparatus for identifying an audio signal|
|US20010043594||May 22, 1997||Nov 22, 2001||Hiroaki Ogawa||Information processing apparatus, information processing method and identification code|
|US20020026343||Aug 9, 2001||Feb 28, 2002||Dennis Duenke||Material and labor cost estimatting method and system|
|US20020097873||Jun 29, 1998||Jul 25, 2002||Rade Petrovic||Apparatus and method for embedding and extracting information in analog signals using replica modulation|
|US20020103883||Jun 19, 1998||Aug 1, 2002||Paul Haverstock||Web server with unique identification of linked objects|
|US20030126445||Dec 20, 2002||Jul 3, 2003||Apple Computer, Inc.||Method and apparatus for copy protection|
|US20030133702||Apr 21, 1999||Jul 17, 2003||Todd R. Collart||System, method and article of manufacturing for authorizing the use of electronic content utilizing a laser-centric medium and a network server|
|US20040028222||Jul 27, 2001||Feb 12, 2004||Sewell Roger Fane||Stegotext encoder and decoder|
|US20040037449||Aug 25, 2003||Feb 26, 2004||Davis Bruce L.||Integrating digital watermarks in multimedia content|
|US20040049695||Dec 24, 2002||Mar 11, 2004||Choi Yang Seo||System for providing a real-time attacking connection traceback using a packet watermark insertion technique and method therefor|
|US20040059918||Dec 15, 2000||Mar 25, 2004||Changsheng Xu||Method and system of digital watermarking for compressed audio|
|US20040083369||Jul 25, 2003||Apr 29, 2004||Ulfar Erlingsson||Systems and methods for transparent configuration authentication of networked devices|
|US20040093521||Jul 11, 2003||May 13, 2004||Ihab Hamadeh||Real-time packet traceback and associated packet marking strategies|
|US20040117628||Sep 5, 2003||Jun 17, 2004||Z4 Technologies, Inc.||Computer readable storage medium for enhancing license compliance of software/digital content including self-activating/self-authenticating software/digital content|
|US20040117664||Sep 5, 2003||Jun 17, 2004||Z4 Technologies, Inc.||Apparatus for establishing a connectivity platform for digital rights management|
|US20040128514||Sep 8, 2003||Jul 1, 2004||Rhoads Geoffrey B.||Method for increasing the functionality of a media player/recorder device or an application program|
|US20040225894||Sep 5, 2003||Nov 11, 2004||Z4 Technologies, Inc.||Hardware based method for digital rights management including self activating/self authentication software|
|US20050160271||Oct 14, 2003||Jul 21, 2005||Brundage Trent J.||Identification document and related methods|
|US20050246554||Apr 30, 2004||Nov 3, 2005||Apple Computer, Inc.||System and method for creating tamper-resistant code|
|US20060005029||Jul 1, 2005||Jan 5, 2006||Verance Corporation||Pre-processed information embedding system|
|US20060013395||May 26, 2005||Jan 19, 2006||Brundage Trent J||Digital watermark key generation|
|US20060013451||Oct 7, 2003||Jan 19, 2006||Koninklijke Philips Electronics, N.V.||Audio data fingerprint searching|
|US20060041753||Aug 11, 2003||Feb 23, 2006||Koninklijke Philips Electronics N.V.||Fingerprint extraction|
|US20070083467||Oct 10, 2005||Apr 12, 2007||Apple Computer, Inc.||Partial encryption techniques for media data|
|US20070127717||Nov 6, 2006||Jun 7, 2007||Juergen Herre||Device and Method for Analyzing an Information Signal|
|US20070253594||Mar 2, 2007||Nov 1, 2007||Vobile, Inc.||Method and system for fingerprinting digital video object based on multiresolution, multirate spatial and temporal signatures|
|EP0372601A1||Nov 8, 1989||Jun 13, 1990||Philips Electronics N.V.||Coder for incorporating extra information in a digital audio signal having a predetermined format, decoder for extracting such extra information from a digital signal, device for recording a digital signal on a record carrier, comprising such a coder, and record carrier obtained by means of such a device|
|EP0565947A1||Mar 31, 1993||Oct 20, 1993||NOKIA TECHNOLOGY GmbH||Procedure for including digital information in an audio signal prior to channel coding|
|EP0581317A2||Jul 30, 1993||Feb 2, 1994||Corbis Corporation||Method and system for digital image signatures|
|EP0649261B1||Oct 17, 1994||Jan 23, 2002||Canon Kabushiki Kaisha||Image data processing and encrypting apparatus|
|EP0651554A1||Oct 25, 1994||May 3, 1995||Eastman Kodak Company||Method and apparatus for the addition and removal of digital watermarks in a hierarchical image storage and retrieval system|
|EP1354276B1||Oct 26, 2001||Dec 12, 2007||Audible Magic Corporation||Method and apparatus for creating a unique audio signature|
|1||Arctic Monkeys (Whatever People Say I Am, That's What I'm Not), Domino Recording Co. Ltd., Pre-Release CD image, 2005, 1 page.|
|2||Bender, et al., Techniques for Data Hiding, IBM Systems Journal, vol. 35, Nos. 3 & 4,1996,pp. 313-336.|
|3||Bender, Walter R. et al., Techniques for Data Hiding, SPIE Int. Soc. Opt. Eng., vol. 2420, pp. 164-173, 1995.|
|4||Boney, et al., Digital Watermarks for Audio Signals, EVSIPCO, 96, pp. 473-480.|
|5||Boney, et al., Digital Watermarks for Audio Signals, Proceedings of the International Conf. on Multimedia Computing and Systems, Jun. 17-23, 1996 Hiroshima, Japan, 0-8186-7436-9196, pp. 473-480.|
|6||Brealy, et al., Principles of Corporate Finance, "Appendix A-Using Option Valuation Models", 1984, pp. 448-449.|
|7||Caronni, Germano, "Assuring Ownership Rights for Digital Images", published proceeds of reliable IT systems, v15 '95, H.H. Bruggemann and W Gerhardt-Hackel (Ed.) Viewing Publishing Company, Germany, 1995.|
|8||Cayre, et al.,"Kerckhoffs-Based Embedding Security Classes for WOA Data Hiding", IEEE Transactions on Information Forensics and Security, vol. 3 No. 1 Mar. 2008, 15 pp.|
|9||Copeland, et al., Real Options:A Practioner's Guide, 2001 pp. 106-107, 201-202, 204-208.|
|10||Cox, et al., Secure Spread Spectrum Watermarking for Multimedia, NEC Research Institude, Techinal Report 95-10, pp. 33.|
|11||Cox, I. J., et al. "Secure Spread Spectrum Watermarking for Multimedia," IEEE Transactions on Image Processing, vol. 6 No. 12, Dec. 1, 1997, pp. 1673-1686.|
|12||Craver, et al., Can Invisible Watermarks Resolve Rightful Ownerships? IBM Research Report, RC 20509 (Jul. 25, 1996) 21 pp.|
|13||Crawford, D.W. "Pricing Network Usage:A Market for Bandwith of Market Communication?" presented MIT Workshop on Internet Economics, Mar. 1995 http://www.press.umich.edu/jep/works/CrawMarket.html on Mar. 12, 1995, cited by other.|
|14||Delaigle, J.-F., et al. "Digital Watermarking," Proceedings of the SPIE, vol. 2659, Feb. 1, 1996, pp. 99-110 (Abstract).|
|15||Dept. of Electrical Engineering, Del Ft University of Technology, Del ft The Netherlands,Cr.C. Langelaar et al.,Copy Protection for Mulitmedia Data based on Labeling Techniques, Jul. 1996, 9 pp.|
|16||European Search Report & European Search Opinion, completed Oct. 15, 2007; authorized officer James Hazel (EP 07 11 2420) (9 pages).|
|17||F. Hartung, et al., Digital Watermarking of Raw and Compressed Video, SPIE vol. 2952, pp. 205-213.|
|18||Fabien A.P. Petitcolas, Ross J. Anderson and Markkus G. Kuhn, "Attacks on Copyright Marking Systems," LNCS, vol. 1525, Apr. 14-17, 1998, pp. 218-238, ISBN: 3-540-65386-4.|
|19||Gerzon, Michael et al., A High Rate Buried Data Channel for Audio CD, presentation notes, Audio Engineering Soc. 94th Convention (1993).|
|20||Gruhl,Daniel et al.,Echo Hiding. In Proceeding of the Workshop on Information Hiding. No. 1174 in Lecture Notes in Computer Science,Cambridge,England (May/Jun. 1996).|
|21||Hartung, et al. "Multimedia Watermarking Techniques", Proceedings of the IEEE, Special Issue, Identification & Protection of Multimedia Information, pp. 1079-1107, Jul. 1999, vol. 87, No. 7, IEEE.|
|22||Horowitz, et al., The Art of Electronics, 2nd Ed., 1989, pp. 7.|
|23||Jayant, N.S. et al., Digital Coding of Waveforms, Prentice Hall Inc., Englewood Cliffs,NJ, pp. 486-509 (1984).|
|24||Johnson, et al., Transform Permuted Watermarking for Copyright Protection of Digital Video, IEEE Globecom 1998, Nov. 8-12, 1998, New York, New York, vol. 2,1998, pp. 684-689 (ISBN 0-7803-4985-7).|
|25||Joseph J.K. O'Ruanaidh and Thierry Pun, "Rotation, Scale and Translation Invariant Digital Image Watermarking", pre-publication, Summer 1997, 4 pages.|
|26||Joseph J.K. O'Ruanaidh and Thierry Pun, "Rotation, Scale and Translation Invariant Digital Image Watermarking", Submitted to Signal Processing, Aug. 21, 1997, 19 pages.|
|27||Kahn, D., The Code Breakers, The MacMillan Company, 1969, pp. xIII, 81-83,513,515,522-526,863.|
|28||Kini, a. et al., "Trust in Electronic Commerce: Definition and Theoretical Considerations", Proceedings of the 31st Hawaii Int'l Conf on System Sciences (Cat. No. 98TB100216), Jan. 6-9, 1998, pp. 51-61, Los.|
|29||Koch, E., et al., Towards Robust and Hidden Image Copyright Labeling, 1995 IEEE Workshop on Nonlinear Signal and Image Processing, Jun. 1, 1995, Neos Marmaras, pp. 4.|
|30||Kocher, et al., "Self Protecting Digital Content", Technical Report from the CRI Content Security Research Initiative, Crytography Research, Inc. 2002-2003, 14 pages.|
|31||Konrad, K. et al., "Trust and Elecronic Commerce-more than a techinal problem," Proceedings of the 18th IEEE Symposium on Reliable Distributed Systems, Oct. 19-22, 1999, pp. 360-365, Lausanne.|
|32||Kutter, Martin et al., Digital Signature of Color Images Using Amplitude Modulation, SPIE-E197, vol. 3022, pp. 518-527.|
|33||Lemma, et al. "Secure Watermark Embedding through Partial Encryption", International Workshop on Digital Watermarking ("IWDW" 2006), Springer Lecture Notes in Computer Science.2006. (to appear) 13 pages.|
|34||Low, S.H., "Equilibrium Allocation and Pricing of Variable Resources Among User-Suppliers", 1988. http://www.citesear.nj.nec.com/366503.html.|
|35||Menezes, Alfred J., Handbook of Applied Crypography, CRC Press, p. 46, 1997.|
|36||Menezes, Alfred J., Handbook of Applied Cryptography, CRC Press, p. 175, 1997.|
|37||Merriam-Webster's Collegiate Dictionary, 10th Ed., Merriam Webster, Inc., p. 207.|
|38||Moskowitz, "What is Acceptable Quality in the Application of Digital Watermarking: Trade-offs of Security, Robustness and Quality", IEEE Computer Society, Proceedings of ITCC 2002, Apr. 10, 2002, pp. 80-84.|
|39||Moskowitz, Bandwith as Currency, IEEE Multimedia, Jan.-Mar. 2003, pp. 14-21.|
|40||Moskowitz, Multimedia Security Technologies for Digital Rights Management, 2006, Academic Press, "Introduction-Digital Rights Management" pp. 3-22.|
|41||Namgoong, H. "An Integrated Approachto Legacy Data for Multimedia Applications", Proceedings of the 23rd EUROMICRO Conference, vol. Issue 1-4, Sep. 1997, pp. 387-391.|
|42||OASIS (Dig Out Your Soul), Big Brother Recordings Ltd, Promotional CD image, 2008, 1 page.|
|43||Oomen,A.W.J. et al., A Variable Bit Rate Buried Data Channel for Compact Disc, J.Audio Eng.Sc.,vol. 43,No. 1/2,pp. 23-28 (1995).|
|44||O'Ruanaidh, et al. Watermarking Digital Images for Copyright Protection, IEEE Proceedings, vol. 143, No. 4, Aug. 1996, pp. 250-256.|
|45||PCT International Search Report, completed Apr. 4, 1997; authorized officer Bernarr Earl Gregory (PCT/US97/00651) (1 page).|
|46||PCT International Search Report, completed Jan. 26, 2001; authorized officer Gilberto Barron (PCT/US00/21189) (3 pages).|
|47||PCT International Search Report, completed Jul. 12, 1999; authorized officer R. Hubeau (PCT/US99/07262) (3 pages).|
|48||PCT International Search Report, completed Jul. 20, 2001; authorized officer A. Sigolo (PCT/US00/18411) (5 pages).|
|49||PCT International Search Report, completed Jun. 11, 1996; authorized officer Salvatore Cangialosi (PCT/US96/10257) (4 pages).|
|50||PCT International Search Report, completed Jun. 30, 2000; authorized officer Paul E. Callahan (PCT/US00/06522) (7 pages).|
|51||PCT International Search Report, completed Mar. 20, 2001; authorized officer P. Corcoran (PCT/US00/33126) (6 pages).|
|52||PCT International Search Report, completed May 6, 1997; authorized officer Salvatore Cangialosi (PCT/US97/00652) (3 pages).|
|53||PCT International Search Report, completed Oct. 23, 1997; authorized officer David Cain (PCT/US97/11455) (1 page).|
|54||PCT International Search Report, completed Sep. 13, 1995; authorized officer Huy D. Vu (PCT/US95/08159) (2 pages).|
|55||PCT International Search Report, date of mailing Mar. 15, 2001; authorized officer Marja Brouwers (PCT/US00/18411) (5 pages).|
|56||Pohlmann, Ken C., Principles of Digital Audio, 2nd Ed., 1991, pp. 1-9,19-25,30-33,41-48,54-57,86-107,375-387.|
|57||Pohlmann, Ken C., Principles of Digital Audio, 3rd Ed., 1995, pp. 32-37, 40-48,138,147-149,332,333,364, 499-501,508-509,564-571.|
|58||PortalPlayer, PP5002 digitalmedia management system-on-chip, May 1, 2003, 4 pp.|
|59||Press, et al., Numerical Recipes In C, Cambridge Univ. Press, 1988, pp. 398-417.|
|60||Puate, Joan et al., Using Fractal Compression Scheme to Embed a Digital Signature into an Image, SPIE-96 Proceedings, vol. 2915, Mar. 1997, pp. 108-118.|
|61||Radiohead ("Hail To The Thief"), EMI Music Group-Capitol, Pre-Release CD image, 2003, 1 page.|
|62||Radiohead ("Hail To The Thief"), EMI Music Group—Capitol, Pre-Release CD image, 2003, 1 page.|
|63||Rivest, et al., "Pay Word and Micromint: Two Simple Micropayment Schemes," MIT Laboratory for Computer Science, Cambridge, MA, May 7, 1996, pp. 1-18.|
|64||Rivest, H. "Chaffing and Winnowing: Confidentiality without Encryption", MIT Lab for Computer Science, http://people.csail.mit.edu/rivest/Chaffing.txt, Apr. 24, 1998, 9 pp.|
|65||Rivest,et al., PayWord and MicroMint: Two simple micropayment schemes, MIT Laboratory for Computer Science, Cambridge, MA 02139, Apr. 27, 2001, pp. 1-18.|
|66||Ross Anderson, "Stretching the Limits of Steganography," LNCS, vol. 1174, May/Jun. 1996, 10 pages, ISBN: 3-540-61996-8.|
|67||Sarkar, M. "An Assessment of Pricing Mechanisms for the Internet-A Regulatory Imperative.", presented MIT Workshop on Internet Economics, Mar. 1995, http://www.press.umich.edu/jep/works/SarkAsses.html on Mar. 12, 1995, cited by other.|
|68||Schneider, M., et al. "Robust Content Based Digital Signature for Image Authentication," Proceedings of the International Conference on Image Processing (IC, Lausanne), Sep. 16-19, 1996, pp. 227-230, IEEE ISBN.|
|69||Schneier, Bruce, Applied Cryptography, 1st Ed., pp. 67-68, 1994.|
|70||Schneier, Bruce, Applied Cryptography, 2nd Ed., John Wiley & Sons, pp. 9-10, 1996.|
|71||Schneier, Bruce, Applied Cryptography, John Wiley & Sons, inc. , New York,1994, pp. 68,69,387-392,1-57,273-275,321-324.|
|72||Schunter, M. et al., "A Status Report on the SEMPER framework for Secure Electronic Commerce", Computer Networks and ISDN Systems, Sep. 30, 1998, pp. 1501-1510, vol. 30, No. 16-18, NL, North Holland.|
|73||Sirbu, M. et al., "Net Bill: An Internet Commerce System Optimized for Network Delivered Services", Digest of Papers of the Computer Society Computer Conference (Spring), Mar. 5, 1995, pp. 20-25, vol. CONF40.|
|74||Sklar,Bernard, Digital Communications, pp. 601-603 (1988).|
|75||Smith, et al. Modulation and Information Hiding in Images, Springer Verlag, 1st Int'l Workshop, Cambridge, UK, May 30-Jun. 1, 1996, pp. 207-227.|
|76||STAIND (The Singles 1996-2006), Warner Music-Atlantic, Pre-Release CD image, 2006, 1 page.|
|77||STAIND (The Singles 1996-2006), Warner Music—Atlantic, Pre-Release CD image, 2006, 1 page.|
|78||Steinauer D. D., et al., "Trust and Traceability in Electronic Commerce", Standard View, Sep. 1997, pp. 118-124, vol. 5 No. 3, ACM, USA.|
|79||Supplementary European Search Report, completed Jun. 27, 2002; authorized officer M. Schoeyer (EP 00 91 9398) (1 page).|
|80||Supplementary European Search Report, completed Mar. 5, 2004; authorized officer J. Hazel (EP 96 91 9405) (1 page).|
|81||Swanson, Mitchell D., et al. Robust Data Hiding for Images, 7th IEEE Digital Signal Processing Workshop, Leon, Norway. Sep. 1-4, 1996, pp. 37-40.|
|82||Swanson, Mitchell D.,et al., Transparent Robust Image Watermarking, Proc. of the 1996 IEEE Int'l Conf. on Image Processing, vol. 111, 1996 , pp. 211-214.|
|83||ten Kate, W. et al., "Digital Audio Carrying Extra Information", IEEE, CH 2847-2/90/0000-1097, (1990).|
|84||Ten Kate,W. et al., A New Surround-Stereo-Surround Coding Techniques, J. Audio Eng.Soc.,vol. 40,No. 5,pp. 376-383 (1992).|
|85||Tirkel,A.Z., A Two-Dimensional Digital Watermark, DICTA '95, Univ. of Queensland, Brisbane, Dec. 5-8, 1995, pp. 7.|
|86||Tirkel,A.Z., Image Watermarking-A Spread Spectrum Application, ISSSTA '96, Sep. 1996, Mainz, German, pp. 6.|
|87||Tomsich, et al., "Towards a secure and de-centralized digital watermarking infrastructure for the protection of Intellectual Property", in Electronic Commerce and Web Technologies, Proceedings (ECWEB 2000), Springer LNCS, vol. 1825, 2000, pp. 38-47.|
|88||Van Schyandel, et al. Towards a Robust Digital Watermark, Second Asain Image Processing Conference, Dec. 6-8, 1995,Singapore, vol. 2,pp. 504-508.|
|89||van Schyndel, et al. A digital Watermark, IEEE Int'l Computer Processing Conference, Austin, TX, Nov. 13-16, 1994, pp. 86-90.|
|90||VeriDisc, "The Search for a Rational Solution to Digital Rights Management (DRM)", http://188.8.131.52/news/whitepaper, /docs/veridisc-white-paper.pdf, 2001, 15 pp.|
|91||VeriDisc, "The Search for a Rational Solution to Digital Rights Management (DRM)", http://184.108.40.206/news/whitepaper, /docs/veridisc—white—paper.pdf, 2001, 15 pp.|
|92||Wayback Machine, dated Jan. 17, 1999, http://web.archive.org/web/19990117020420/http://www.netzero.com/, accessed on Feb. 19, 2008.|
|93||Wong, Ping Wah. "A Public Key Watermark for Image Verification and Authentication," IEEE International Conference on Image Processing, vol. 1, Oct. 4-7, 1998, pp. 455-459.|
|94||Zhao, Jian et al., Embedding Robust Labels into Images for Copyright Protection, (xp 000571976), pp. 242-251, 1995.|
|95||Zhao, Jian et al., Embedding Robust Labels into Images for Copyright Protection, Proceeding of the Know Right '95 Conference, pp. 242-251.|
|96||Zhao, Jian. "A WWW Service to Embed and Prove Digital Copyright Watermarks", Proc. of the european conf. on Mulitmedia Applications, Services & Techniques Louvain-La-Nevve, Belgium, May 1996.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7970722||Nov 9, 2009||Jun 28, 2011||Aloft Media, Llc||System, method and computer program product for a collaborative decision platform|
|US8005777||Jul 27, 2010||Aug 23, 2011||Aloft Media, Llc||System, method and computer program product for a collaborative decision platform|
|US8160988||Jul 27, 2010||Apr 17, 2012||Aloft Media, Llc||System, method and computer program product for a collaborative decision platform|
|U.S. Classification||380/28, 382/181, 713/176|
|International Classification||H04L9/00, G06K9/00|
|Cooperative Classification||G06T2201/0064, G06T1/0028|
|Sep 7, 2007||AS||Assignment|
Owner name: WISTARIA TRADING, INC., FLORIDA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:MOSKOWITZ, SCOTT;REEL/FRAME:019859/0593
Effective date: 20070815
|Mar 8, 2011||CC||Certificate of correction|
|Nov 29, 2013||FPAY||Fee payment|
Year of fee payment: 4
|Aug 18, 2015||AS||Assignment|
Owner name: WISTARIA TRADING LTD, BERMUDA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:WISTARIA TRADING, INC.;REEL/FRAME:036387/0177
Effective date: 20150814